The HTC Weekly #2

Social Media in Chaos!

Jan 19, 2025

The Creative in a Time of Connectivity

It is a paradoxical time to be a creative, both difficult and exciting. On the one hand there are more opportunities than perhaps have ever existed in earlier eras; opportunities to create, for the creations to be seen, and to connect with an audience. On the other hand, the pitfalls of modern connectivity and Life Under The Algorithm can be a subtle trap for anyone hoping to make a life from their own creativity.

This is not my first time publishing writings on the internet. During the mid-2000’s I published a modest technical blog catering to developers and techs. In a time before the ubiquitous developer help of Stack Overflow and Reddit, it was sites like mine that served to share answers between developers in the online community. The proper way to interpret obscure error messages, the way to fix obscure bugs, the dangers of a particular update - this was the content we shared amongst ourselves.

It was an exciting time. The e-book revolution had yet to start but it existed in early stirrings through early adopters outfitted with second hand PDA’s and a working knowledge of Project Gutenberg. The democratisation of audiences was just beginning; no longer was it necessary for a publisher gatekeeper to approve of you for your writing or your art to exist in a publicly accessible space; anyone could simply publish to the internet. The barrier of entry was enormously low and if there was no built in audience for a new site at the very least it was there, for people to see. Content discovery was a concern in these early days but somehow we managed through early forums and chat relays, word of mouth having always had wings.

This was the start of the podcast age as well. There is a mistaken belief that podcasting is somehow a recent phenomenon, perhaps beginning with the explosive success of Serial, and the wildly successful D&D-branded podcasts of the mid 2010’s. The early-mid 2000’s though were when audio producers, often with just a cheap microphone and production values that would cause a modern listener’s ears to bleed, realised that they too no longer needed to placate a gatekeeper. No longer was the barrier of entry the few remaining broadcast radio stations, the handful of audio-drama stations left in the world for instance (the BBC springs to mind). Instead, anyone could put together a script, gather a cast, and create and publish.

Occasionally a modern podcaster will complain about the crowded marketplace now, the difficulty of getting noticed, and refer to themselves as “having been around before everyone jumped in on it” despite dating to the mid 2010’s themselves. Even some of my favourites have been guilty of this little conceit. (I’m looking at you James and Jimmie). RPGMP3 released some of the first podcast actual-plays in 2004, recorded on a crappy home microphone recording people around a dinner table. Podiobooks began serialising audiobooks in podcast form in 2005. None of this is new.

What has happened since, then? In two words; Social Media.

Social Media is a two edged sword for creatives in the modern age. On the one hand it offers an audience. When the gatekeepers were no longer needed the big problem was discoverability; The benefit of the old publishing model, of newspapers, magazines and book publishers, was a built in audience. You needed to impress an editor for an article to appear in a magazine but when it did all the magazine subscribers would see and read it. On the internet there was no blocker to you publishing what and when you wanted but building an audience was entirely up to you, and it was (and still is) a difficult task. Some manage, and do extremely well. Many fail.

Social Media offered another way. The same lack of gatekeepers but with a built in audience, all being trained by the new platforms to share and comment, to like and subscribe. It seemed too good to be true: and of course it was. In order to access this audience, creatives came to realise they were now under the control of a new and more merciless gatekeeper than ever before: The Algorithm.

As a tech, I find the various algorithms used by social media companies fascinating, particular those that work well. It’s also worth saying that I found the experiments performed by Meta on whether they could affect their users moods by adjusting the algorithm that controlled their feeds fascinating as well - it was still a reprehensible thing for them to do.

TikTok, to my mind, has had the most impressive algorithm so far. As a husband and father to a teenage daughter and son, I have had a chance to witness TikTok through the lens of four very different people,and I was surprised (and impressed) to find that each of us had extremely different experiences on the platform. Different to the extent that we could well have been looking at entirely different platforms for all that the content had in common. My wife’s algorithm is strange to me, a mix of emotional content and crafts. Strange funny memes mix with a fetish tailor who makes custom latex bodysuits. It was chaotic and engaging. My son’s algorithm; cat videos and video game walk-throughs narrated by high-energy streamers that fill me with rage just to hear them.

My daughters algorithm? Music, dance trends, and whatever is in with the teenagers at the moment.

Mine? Almost entirely comedic or artistic. Videos of people restoring old tools by hand mix with painters demonstrating their techniques mix with a one-woman sex-positive narrative show about a help-desk in hell. The darkest thing on my feed was a series of videos made by a mentally ill artist who makes ink-drop drawings of creatures from Nordic mythology. He’s incredible. More common on feed though were videos of a gen-x’er who liked to respond to the ridiculous things gen z’s say with gentle rants and explanations of how the world used to be.

This custom curation comes at a cost however: if you don’t understand the algorithm, you may find it hard to satisfy it. The algorithm is your gatekeeper now, deciding whether your art, or writing, or video reaches an audience or remains cold and unloved on your own feed.

Many have turned to “understanding” the algorithms to try and trick it into “making them viral” and increasing their popularity. No doubt sometimes this works. Algorithm experts have popped up, much like SEO experts did when it became clear that high-ranking on google was worth a great deal of money to the right people.

People have made entire careers out of teaching others how to manipulate the system to try and get ahead, whilst others just hope for the best (to quote my favourite podcasters “If you want to help, please review us and leave 5 stars. It’s not for our ego, it really does help. We don’t understand why, it’s the algorithm, just please, it helps”).

This environment has given rise to the “content creator”. Not quite an artist or creative, not quite a technician or business specialist, but something in the middle. They create like an artist, but what they create is driven by trends or what they think will appeal to the algorithm. Once again we find ourselves back in the old days of publishing, trying to convince an editor to believe in us, but this time the editor is all code and statistics and completely impervious to charm or persuasion.

Some have done very well by it. I walked away for a long time for while I loved the art. I loved the writing, the drawing, the creating, I found appeasing the algorithm and building the audience mind-numbingly dull. Many others have found the same, or have tried to walk the path and found themselves unable to put in the work, the time, the stress, that can be needed to reach sustainability. For every overnight viral success there are a hundred who burn out before they make it far enough to feel proud of themselves.

So what is a creative to do in this environment? When the world is connected and the barrier to entry is lower than ever, but the algorithm drives all attention, what do you do?

The answer, it seems, is really quite simple. Do what creatives have done since the dawn of time and create. Don’t spend your time worrying about trends, SEO filtering, engagement scores and the right way to share in order to maximise hits. Unless that is what gets your juices flowing, in which case go nuts and more power to you.

What is important though is the act of creation, and doing so in a way that makes you personally fulfilled. It doesn’t matter if you create pen and ink drawings, write erotic fiction, create home-made cooking videos, or build generative art workflows to create anime-style landscapes; If you find the thing that makes you excited to create you can carry on for as long as you need to, trying to build an audience. If you’re creating in a space you love, a space you need to be in, then you can survive the long nights when your growth is stagnant and it feels like no-one is listening. Cultivate creation as its own reward and dedicate yourself to creating what gives you joy to create and you are far more likely to bring to life a work of art that means something, that resonates, and that finds the audience that needs it, than if you chase every trend, creating work that leaves you cold but is designed to “go viral”.

This isn’t to say that social media isn’t important. It really is a built in audience, but you ideally want to use it to find the audience that is looking for you and what you have to say, rather than trying to alter your message to fit what you think a specific audience wants.

Both can work of course, and there are plenty of people, commercial artists and writers, who will tell you the exact opposite, many of them far more successful than I ever hope to be. But when I look at another abandoned blog, another Instagram feed with a half-dozen posts, another social media engagement attempt that fizzled out quickly… I often wonder if they may have hung in longer, long enough to build something perhaps, if they had enjoyed what they were doing more. If they were doing what they wanted to do, what their muse needed them to do, rather than what they thought they had to do.

We do all need to eat sadly, there are commercial considerations to every decision including what to create. Perhaps my advice here is better suited to weekend creators, those with day jobs rather than those trying to create for a living. Certainly over the decades I have built many useful but unfulfilling pieces of software and infrastructure and made a great deal of money doing it, perhaps that is the nature of the modern world. I guess what I want to know is: does it have to be?

There are opportunities all around us, strange ones sometimes. For instance, Google has started paying video creators for their unused footage; any scraps or out-takes they may have on their hard drive, footage that was edited out of completed work. Why? Training fodder for their video AI models. That’s an opportunity that didn’t exist not long ago.

So, what’s the takeaway for creatives in this brave new world? Don’t let the allure of viral fame or the siren song of the algorithm drown out your own creative voice. In a time of generative content and flash-in-the-pan trends, it has never been more important to be true to yourself. Embrace the opportunities all around us but remember that your unique perspective and artistic integrity are your most valuable assets. Cultivate a practice that nourishes your creative passion, and let that passion be your guiding star. When you create from a place of authenticity, you'll attract the audience that resonates with your work, the ones who crave the very essence of what makes you, you. In a world increasingly driven by data, let your art be the human touch that reminds us all of the enduring power of creativity.

Nick Bronson, Editor-in-chief

Further Reading:

The Headline Hamster Wheel

News updates to previous weeks articles, as the news continues!

Update: The A.I. Disruption

Some good news this week as further analysis of the World Economic Forum (WEF) Future Jobs Report showed that although they predicted A.I. to cost traditional industries 92 million jobs globally by 2030, they also expected it to create a further 171 million new jobs in both A.I. development industries and in traditional industries working with the new A.I. models.

The WEF is predicting a surge in demand for professionals with expertise in AI/big data analytics, cybersecurity and general technical literacy. However some traditional industries, particularly postal clerks, executive & legal secretaries, payroll staff and, for the first time, graphic designers are all expected to suffer a contraction in the near future.

Despite this news we have been reminded quite forcefully that the short term is going to be bumpy, as both Meta and Microsoft have announced further job cuts to come, but in the name of “cost-control” and “poor performance”; which appear to be the common code at the moment for A.I. related retrenchment. Meta intends to lay off a further 5% of its staff across the business as it pivots hard towards A.I. with Microsoft following a similar line.

The key takeaway from this appears to remain as it was previously; the smart investment to make in yourself right now, particularly if you are in a clerical back or mid-office job that could be in danger of disruption, is to start learning and upskilling, particularly in A.I. related domains.

In the meantime a tech startup out of Florida, Mega HR, launched an agent service this week called “Megan”, a A.I. powered human resources agent that the company claim can automate most recruiting and hiring tasks while improving communication with job applicants.

Time will tell if Megan can assist in the hiring process, though from the horror stories it’s unlikely she could possibly make things much worse.

Further Reading:

Update: U.S. A.I. Protectionism

News has continued to filter out this week about the world’s response to the Biden administration’s last-ditch attempt at regulating the global A.I. industry, and outside of the U.S. the response has been lukewarm at best.

Much has been made this week of potential changes to the landscape when Trump takes office on the 20th January. We are still waiting for further clarity on Trump’s stance on A.I. but analysts have pointed to several likely scenarios based on what we know of his policies in general.

In foreign policy, Trump is known for differing from the last administration on a number of points. Though he has been a staunch opponent of China and is unlikely to directly relax restrictions there, he has historically been far friendlier towards Russia. Given the closer relations that China and Russia have been enjoying, this could be good news for China should Trump be convinced to relax some restrictions to Russia. A further surprise to some was the Biden administration’s inclusion of Israel on the restricted nations list, likely due to the ongoing troubles in Gaza. Given Trump’s outspoken support of both Israel and Netanyahu, it’s not out of the realm of possibility this may change.

Trump is also known for having a business friendly stance in general and counts tech billionaire Elon Musk amongst his closest advisers. He is also known for his love of tariffs and protectionism however and the idea of exercising control over the world market for A.I. chips is likely one that will resonate with him personally. There are going to be significant competing pressures on him which further complicates any prediction of how he will decide. Trump’s A.I. Czar, David Sacks, is a Silicon Valley entrepreneur and the Republican party platform itself promised to repeal Biden’s A.I. executive order, so it does seem likely that at least some deregulation can be expected.

The European Union (EU) meanwhile has raised concerns about the restriction’s effects on its member states, as not every member of the EU was granted membership in the top tier, unrestricted list. They have reiterated that they see their relationship with the US as an opportunity, not a security risk, however given some EU member’s ambiguous attitudes towards Russia and China it seems unlikely that a unilateral move will be made there.

The tech industry, including Nvidia and Microsoft, have continued their objections to the rules on the grounds that it will harm competitiveness and innovation in the A.I. space by targeting tech that is already widely available, such as GPUs; seeming to suggest that the rules could be applied to GPU’s used in gaming PCs as well as those specifically designed for data center AI racks.

The next few weeks will be critical to determine how much of the Biden restrictions will continue and what affect they will have on the global marketplace. A lot of eyes (and lobbyists) will be watching Trump in his first few weeks waiting to see.

Further Reading:

Update: TikTok: If This is the Last Dance, then save it for me baby…

[Further Update: Since this update was written the supreme court has returned with the decision to uphold the ban. Late on Saturday the 18th in the U.S., ahead of the ban officially beginning on the 19th, TikTok shut down their U.S. operations and “went dark.” Time, and likely Trump, will tell if this is a temporary measure or something long term. -Ed]

This week it has often felt as if the world could talk about little else but TikTok and its upcoming fate; something of a frustration when there has been so little real additional information to hand.

The Supreme Court decision on whether or not to uphold the ban is imminent, expected by January 18th (likely not long before this article is due to go to press). If the ban is upheld, which most roundly expect it to be, the ban will take place from the next day, January 19th.

The immediate stakes are high with the Supreme Court essentially able to offer a stay of execution by striking down the law and allowing TikTok to continue to operate unhindered. However this ruling is widely seen as a test case for upholding government national security concerns against a citizen’s constitutional right to freedom of speech and assembly. Given the current makeup of the court and its past decisions, few expect this court to rule in favour of the first amendment.

There has been significant movement in the political space around the upcoming ban however competing agendas have essentially gridlocked any chance of change coming from that arena. Biden, just days away from the end of his term, has adopted a hands off stance and left the issue to the incoming administration to sort out. Senator Markey introduced a bill to extend the deadline by 275 days in order to examine and facilitate other solutions on Jan 14th, however the bill was blocked by senate republicans and unable to pass.

TikTok itself has announced that despite some commentators speculating that they may try to “limp” along after the ban by continuing to provide service to users who already possess the app (the law will require app stores to cease offering it, but not users to delete it), it is not interested in “carrying on crippled” and if the ban was upheld, they plan to shut down services to the U.S. immediately on the 19th of January.`

TikTok highlighted their commitment both to user data privacy and to user agency by stating that all users will be notified before the shutdown and able to download copies of all of their data stored on the servers. In addition they announced that regardless of the outcome, the U.S. employees of the company did not have to fear for their jobs, they would still be employed post-ban.

There is a deep irony that a company whose work is being banned under the pretext of national security is showing so much concern for both its U.S. customers and employees at a time where the rest of the U.S. tech industry is determined to squeeze the life out of both.

Rumours earlier in the week speculated that China was considering allowing the divestment of TikTok after all, potentially to Elon Musk. There could be some political capital to be made there as Musk is extremely close to president Trump and relations between Trump and China are expected to be fraught, at least initially.

As the week has progressed however this has seemed increasingly unlikely given the regulatory hurdles and the time remaining. Many would be unhappy to see further consolidation of the social media landscape (Elon Musk already owns Twitter, now X, as a privately owned concern) and there are those who would worry that the data represented by TikTok’s databases would be no safer in Musk’s hands than China’s.

At this stage the most likely scenario appears to be a shutdown of TikTok in the U.S. mirroring India’s decision several years ago.

In the meantime an exodus of users from TikTok, keen to save their livelihood, have spread out to a number of existing mainstream and non-mainstream social media platforms. The big winners this week appear to be RedNote and Lemon8, ironically two social media companies also owned by Chinese companies; Lemon8 being owned by the same parent company that owns TikTok! Speculation over whether this growth is going to spur the U.S. to ban these networks as well is rife; the Trump administration is likely to find themselves in a difficult spot here. Banning these networks is going to be extremely unpopular with the users, particularly those who have just now fled the TikTok shutdown, however not applying the same rules to these likely renders the banning of TikTok completely pointless as China has exactly the same potential access to data owned by RedNote or Lemon8 as they did TikTok.

Other networks such as Instagram and YouTube Shorts have likewise seen a growth in user numbers during this time. More niche open social networks such as Bluesky on decentralised platforms have also seen a significant bump in users lately, likely from those wishing to absent themselves from the Musk & Zuckerberg controlled walled gardens after both have made large policy pivots aimed at getting closer to the incoming administration this week.

Regardless of how the Supreme Court rules, the next few days are going to be crucial not only for TikTok but for the tech landscape as a whole; setting legal precedent, rearranging user bases and, no doubt, causing an awful lot of political angst.

Further Reading:

Update: Corporations Behaving Badly: I never Meta AI I didn’t like

A small update on the Kadrey v. Meta Platforms, Inc. copyright legal battle this week as the plaintiffs lodged an amended allegation claim and further redacted transcripts of trial depositions have been released to the public.

The big news is the amended allegations where the plaintiffs claim that not only did Meta’s A.I. teams train their models on books they knew to be pirated through the LibGen repository, but they cross referenced books available in that repository with books available for licensing in order to determine which works they needed to license and which they were able to avoid licensing by pirating instead. If proved, this makes it quite difficult for Meta to claim any sort of ignorance defence about the need to license, or a misunderstanding about the copyright status of texts in the LibGen repository.

According to transcript releases, Mark Zuckerberg’s defence has been the rather disingenuous “I don’t really know what LibGen is” despite internal documents showing he directly approved the use. Further claims in the update allegations include that Meta also used pirated data from the repository “z-library” as recently as April 2024 in its model training and that it seeded the training data with custom entries specifically designed to hide the fact that pirated data had been used to train it.

Further Reading:

In AI copyright case, Zuckerberg turns to YouTube for his defense

News of the Week

The New Emperors: Big Tech and the new American Oligarchy

“In the councils of government, we must guard against the acquisition of unwarranted influence, whether sought or unsought, by the military-industrial complex. The potential for the disastrous rise of misplaced power exists and will persist” - Dwight D. Eisenhower

“Americans are being buried under an avalanche of misinformation and disinformation, enabling the abuse of power.” - Joe Biden

In 1961, Dwight D. Eisenhower delivered a farewell address to the nation in which he warned against the dangers of the military-industrial complex - embedded in the government, fatted on government contracts and rich with cold-war funding, Eisenhower believed this group of businesses carried far more influence over the workings of government than they should and this had the potential to cause great harm to the American people should their welfare and the desires of industry cease to coincide - a distinct possibility with an industry that profited greatly from warfare without itself paying the horrific costs.

The parallels have not been missed as Joe Biden gave his farewell speech this week, echoing Eisenhower’s warning as he talked about the dangers of a new unelected oligarchy seeking to rule the country and an unrestrained tech-industrial complex.

“Today, an oligarchy is taking shape in America of extreme wealth, power, and influence that literally threatens our entire democracy, our basic rights and freedoms, and a fair shot for everyone to get ahead.” - Joe Biden

The oligarchs in question weren’t named but there doesn’t seem to be a lot of confusion as to who he was referring to; the tech billionaires occupy pop culture attention in much the way movie-stars and celebrities did in times past though with less grace and far less charisma.

A list of Trumps inauguration donors might be a good place to start in assembling a list of possible oligarchy members. The incoming administration has raise a record-setting $200m for the inauguration, not counting a party due to be thrown by Zuckerberg afterwards. This far outstrips Biden’s inauguration at $61.8M and even Trumps first term inauguration which raised $106.8M.

A significant amount was donated by big tech, with Mark Zuckerberg (Meta), Jeff Bezos (Amazon), Tim Cook (Apple), Sam Altman (OpenAI), Sundar Pichai (Google), and Satya Nadella (Microsoft) all donating $1M each and each planning to attend the inauguration in person. They will be joined by Elon Musk who surpassed them all with his $250M in donations towards helping Trump win the election, for which he has been given a close advisory position to the president. Quite a line-up of potential oligarchs.

Democratic Senators Elizabeth Warren and Michael Bennet have put pressure on these tech luminaries by formally requesting in writing that they explain their donations, suggesting that they are attempting “to avoid scrutiny, limit regulation and buy favour”. It would be disingenuous in the extreme to suggest these donations were given purely out of civic duty however few expect Warren and Bennet to have much effect.

Kiss the Ring

Anyone watching tech this week has been treated to a display of deference rarely seen outside of a Godfather film as these would-be oligarchs have lined up and tripped over themselves to endear themselves to the incoming administration. Trump and his people are known for taking perceived slights very personally and are likely carrying left-over resentment from his previous administration. It is widely agreed that, given his participation in the election (and the amount of money spent), Elon Musk is far and away the front-runner in terms of endearing himself to Donald Trump but the others are not planning on being left behind.

Much to the displeasure of reporting staff at the Washington Post, Jeff Bezos has made a start by interfering there. Trump is known for his disdain for the “mainstream media” and anyone who prefers actual facts over his convenient explanations. According to insiders Bezos intervened directly to prevent the paper from endorsing Kamala Harris prior to the election, ostensibly as the paper “shouldn’t be endorsing any political position”, a policy that may have been more believable had they not now spent the last week specifically endorsing Trump’s selections for his cabinet.

Additionally insiders have claimed that the Post has been shopping around a new slogan internally as a replacement for its current one. The current slogan was adopted in 2017 during Trump’s first term as a direct response and challenge to the perceived misinformation of the time; “Democracy Dies in Darkness”. This slogan was widely taken as a call to action, a pledge to stand up and report truth in the face of pressure.

The new slogan, allegedly, is “Riveting Storytelling for All of America”; a fairly telling change of direction for the paper and it isn’t difficult to come up with potential reasons as to why that might be. When reached for comment by several media outlets the Post has denied that it is planning to change its slogan.

Since Bezos’ interference has become public knowledge the paper has lost a quarter of a million subscribers and reportedly now has less than 3M digital subscribers overall - a far cry from the “200M paying users” they aspire to. Sadly it would seem that the paper best known for standing up to corrupt administrations and daring their displeasure in order to publish the Watergate reports has decided on a different editorial stance than in times past.

The Washington Post is not Bezos’ only attempt at ingratiating himself with the new administration; as well as the donation to his inauguration fund, Amazon is reported to have spent $40M securing the rights to broadcast a documentary about Melanie Trump, something the world was definitely crying out for.

In the meantime, in another sop to the ongoing culture wars, OpenAI has been quietly making modifications to its “economic blueprint for A.I.” in America, removing language that stated that A.I. should “aim to be politically unbiased by default.”

Bias in A.I.; and in the greater world and media landscape in general; has been a popular right-wing trope for many years now. Elon Musk and new Trump A.I. Czar David Sacks have both accused AI chatbots of censoring conservative viewpoints, singling out ChatGPT in particular for being “programmed to be woke” and untruthful about politically sensitive subjects.

Elon Must has mused that this is due to both the data that A.I. is trained on and the “wokeness” of the San Francisco bay area firms, building woke nihilistic philosophy directly into A.I. Academic research has seemed to confirm these ideas, suggesting that ChatGPT has a liberal bias on topics such as immigration, climate change and same-sex marriage.

Of course, others would argue that this is to be expected: these topics are generally replete with scientific facts that the right-wind do not wish to accept. Regardless of their preferences, Climate Change is an accepted fact within scientific circles and easily measurable; in fact it has now progressed so far that the effects on weather patterns do not require statistical analysis, they are easily visible in the fires sweeping Los Angeles and Australia and the rising sea-levels threatening to destroy island nations in the South Pacific; and years of research have found no credible evidence that there are any reasons to deprive same-sex couples of all the same rights as heterosexual couples (the old saws about paedophilia and poor outcomes for children raised by such couples having been well and truly debunked), likewise the tired old claims of climbing crime due to immigration (as well as more recent claims of pet-consumption) are not supported by statistical evidence. Any model trained purely on accepted facts would be bound to disagree with the average right-wing pundit on these topics and bias has less to do with it than wilful ignorance.

Thus that old joke, “reality has a well known liberal bias”.

Not to be outdone by Bezos or Altman, Zuckerberg has also had one busy week this week. The press reports about Meta have been fast and thick on the ground as Zuckerberg appears to be looking to embrace his “Elon Musk” era.

The news out of Meta began with word that they would be discontinuing their fact checking program. This has been a perennial target of right-wing diatribes, which is unsurprising for a group who never likes to let an inconvenient fact get in the way of a good story. Under the new rules Meta will be implementing a “community notes” system similar to what is used in X under Musk and all moderation policies are being relaxed.

In order to make sure everyone fully understood his reasons for these changes Zuckerberg spent 3 hours on Joe Rogan’s podcast this week in one of the more blatantly revisionist interviews ever given to a public audience. Amongst the claims Zuckerberg made were:

- The Biden administration forced Facebook to take down things that were actually true around COVID, under the guise of Covid-19 misinformation.

- Fact checkers were politically biased.

- That social media, particularly Facebook, had no effect on the 2016 election and he regrets accepting the public narrative that it did.

- That the fact checking policy was analogous to something out of Orwell’s 1984.

If these claims weren’t enough to make your head spin, he then justified changes to the moderation policies by claiming that under the policies as they had existed, it would have been impossible to have a debate on whether or not women should serve in combat roles in the military - an opinion (that they shouldn’t) that is publicly held by Trump’s current pick to run the Pentagon, Pete Hegsethe. Zuckerberg’s claim is that any discussion that can be had on the floor of the congress should be acceptable for social media.

He may even have a point there; though why it is necessary to relax moderation rules to permit the denigration of women, immigrants and LGBTQ+ people, permitting them to be compared to objects, called property, insulted or have violence incited against them in order for us to have that debate isn’t entirely clear.

In perhaps the most disingenuous claim made on the podcast, Zuckerberg claimed to have started his work in social media “to give the people a voice”, obviously in the hope that the world has forgotten that he quite famously originally created Facebook as a rather disgusting tool to allow students at Harvard to rank women on campus by their perceived attractiveness. Given the current trend on women’s rights favoured by the right, perhaps that is what they would consider “giving the people a voice.”. Zuckerberg’s rather transparent grasping for “freedom of speech” credibility, at least amongst those who share his views, puts him in good company with Elon Musk, another censor masquerading as a free speech crusader.

Even before we had time to fully process the removal of fact checking and moderation from Meta’s systems however this was immediately followed up with an announcement that Meta was ending all Diversity, Equity and Inclusion programs (DEI). DEI programs are a particularly bugbear of the right who claim that by attempting to address systemic bias they unfairly prejudice systems against… well, them.

Meta’s internal memo announcing the DEI changes state that they will be making changes to their “hiring, development and procurement policies” due to “changes in the legal and policy landscape surrounding DEI efforts in the United States.”. It can hardly be telegraphed more clearly than that.

As a last little twist of the knife, Meta followed up this announcement by quietly removing both the Trans and the Non-binary support themes from Facebook messenger. Not only that but going as far as to removing even the original posts that announced their existence originally from the site; proving that even if Zuckerberg isn’t capable of using Orwell’s 1984 as a credible analogy, he certainly understood the processes employed by the Ministry of Truth well enough.

Not to be outdone, Amazon quietly announced later in the week that it too would be ending DEI programs, though in not quite as proud and definitive a manner. Additionally it modified it’s “Our Positions” policies page which previously contained defined positions on a variety of civil rights issues, including “Equity for Black People”, DEI and LGBTQ+ rights. These have all be replaced with a generic section claiming Amazon believes that the inequitable treatment of anyone is unacceptable. I guess for Amazon, “All Lives Matter”.

Quid Pro Quo

It seems obvious that many of these changes are aimed at pleasing the new administration; which then begs the question, what do these tech oligarchs seek to gain? Trump’s quid-pro-quo approach to business is well documented at this stage, it seems reasonable that these large tech companies expect to receive something in exchange for their donations and policy shifts.

Elon Musk has embedded himself in the administration itself but for the others we are waiting for some action from the companies as to show what their strategy might be. There are some signs however.

OpenAI has opened the year by trying to help steer the regulatory and business environment around A.I. in general by releasing the “Economic Blueprint for A.I.” for the incoming administration, providing policy recommendations on a variety of A.I. related topics.

The blueprint encourages the government to adopt policies that can facilitate additional infrastructure investment in order to create the large data centres that are going to be required for advanced A.I. training. The Biden administration made some movements in this direction and OpenAI would be hoping to convince the Trump administration to continue, and grow, investment in this area.

The blueprint also contains a call to the government to allow significant investment in A.I. projects by investors from the middle-east, something the Biden administration was very wary of. The central claim is that this money will flow to China if investment isn’t arranged with the U.S. Rather than viewing nations like the United Arab Emirates and Saudi Arabia as allies, the blueprint compares them to swing-states in an election, with the U.S. or China as the beneficiaries of their investment.

The blueprint, unsurprisingly, calls on the government to take a light hand in A.I regulation and safety measures, claiming that strong regulations will create dissonance on the national security and economic competitiveness front.

Somewhat controversially, OpenAI is also calling for all publicly available data to be made available to train A.I. models regardless of copyright or IP status. A self-serving request, but one with an interesting rationale in this case; the data, being publicly available, is going to be used for this purpose regardless of U.S. laws, claims OpenAI. The U.S. legal system can restrict U.S. companies from doing so, or require them to pay punishing licence fees for the privilege, but they will be unable to do the same to other actors; including foreign nations. Countries such as China will not respect IP held by Americans, goes the argument, so this training is going to happen - and if it is prevented within the U.S. then they will fall behind other countries who are less ethical in what they choose to use.

Self-serving, certainly, however OpenAI’s analysis is likely not incorrect here either and this may be the best argument raised so far for a blanket allowance on training data, utilitarian and nationalistic it might be.

Zuckerberg is the only other oligarch tipping his hand at the moment, in small way at least, calling on the Trump administration to stop the EU from levying fines on US companies. Given the importance of relations with the EU, it seems unlikely that Trump will seriously consider this step, however it is one that is tailored to appeal to his desire to assert US power in the economic arena; and in order to make it appeal more Zuckerberg likened these fines to “A tariff”, an obsession of the incoming president’s.

This request is likewise self-serving; Meta has been charged more than $2.67B in fines over the last several years for EU data protection regulation violations alone, and an additional $813.71M fine for violating EU anti-trust rules.

A Trick of the Light

There is more at stake here however than simple regulation manipulation and fine avoidance; an excellent think-piece in the bigthink this week laid it out as “ability computing”.

The prediction put forward in the think piece is that by 2030 we are likely to have developed “superhuman” abilities. A stunningly click-baitable headline but not quite as unbelievable as it sounds; the basic argument is that as augmented reality devices are combined with context-aware generative A.I. over the next few years we are going to have access to tools so ubiquitous and useful that they will feel more like extensions of our own abilities rather than tools we are using.

To paraphrase an example from the article, imagine walking into an office in 2030. You are wearing augmented reality glasses connected to your phone and a context aware A.I. that has access to the same perceptual reality as you do. It can see your surroundings via cameras mounted on your glasses or earphones, microphones monitor the surrounding soundscape, the A.I. is able to detect where your attention is focused and offer contextual information by predicting what you might need before you even think to ask.

A man approaches you in the office and your A.I. companion senses your discomfit, accurately determining that you are trying to remember his name. Your augmented reality glasses immediately provide you the information, his name, role and basic data, smoothing out the interaction to come. He speaks to you, in French, and in your ear a voice smoothly translates to English. Later as you walk down a street you wonder out loud when a neighbouring bakery opens and are immediately provided with the answer.

Over time this sort of ultra-connectedness would become second nature, much as smart phones changed our expectations with regards to ubiquitous access to information and communication; in fact you will be at a significant social and commercial disadvantage if you choose not to partake in the technology, likely leading to wide-scale uptake. This is one future that some companies, such as Google, have been chasing for a significant amount of time and resources and as the technology advances it isn’t unreasonable to believe that the time might be close at hand.

A fascinating idea, explored in many works of science fiction and the day-dreams of many tech utopians, so what is the problem? The problem is, of course, exactly what Biden is trying to warn us about: oligarchical control.

The basic concern has been explored in many media; one that lingers in the mind was a pixel-art adventure game entitled “Virtuaverse”, with difficult puzzles and a score by chiptune master “Master Boot Record”. One of the core conceits of the game was that the main character had not followed along with the majority of the world’s population and gotten “implanted” with a virtual overlay that allowed him to interact with the augmented reality world that surrounded him, and so must make do with augmented reality glasses instead. The benefit of this to both gameplay and narrative is that you can choose to wear these glasses or take them off - allowing you to see the world as it is both with and without the augmented overlay; and there is a significant difference. It’s impossible to interact with the modern world without being able to see the neon signs, controls and overlay icons that make up the augmented layer, however with the layer disabled it is easy to see just how much is covered up by it. The world looks grimmer, dirtier, and so much is hidden by the garish lights of the virtual overlay.

Most of the large A.I. models right now are run by the large tech companies. OpenAI (ChatGPT), Microsoft (Copilot on GPT tech), Google (Gemini), Meta (LlaMA), Anthropic (Claude) - all big tech companies offering their A.I. via a service. If these A.I.’s are the engines that drive augmented reality that means that the reality you see once you are embedded in that ecosystem is manipulable by those companies. Whether it’s an advert overlaid on a billboard, a question answered with a particular slant, a translation given a different twist; there are a million ways to subtly alter how we perceive reality once we offload part of our interpretation to an external actor. If this future was ever to become reality - and several large tech companies (Google & Meta for instance) are very keen that it does - then the companies most practised at influence and manipulation on a mass scale will have unmatched access to shape your very perception of the world. Arguably, a deeper version of what Meta and Google do every day right now.

This is why Biden and others are and will continue to sound the alarm about the tech oligarchs and their plans for us; and its why that alternatives like open-source A.I. is more important than ever. If these technologies are going to power the future, not only the future of our society but the future of what we are as people, they cannot be allowed to be controlled by a mere few untrustworthy hands.

It’s possible at this stage you are thinking this is a lot of alarmist hand wringing about unlikely futures. With that in mind, here is some more news from the week.

An article debunking an anti-net-neutrality study disappeared from the internet this week, prompting it’s author to once again ask an existential question of the modern age - If it’s on the net, but Google doesn’t list it, does it still exist?

An article providing a deep dive into the flaws of a study commonly cited by anti-net-neutrality lobbyists and telecoms companies was delisted on major search engines and social networks this week, essentially removing it from the internet (or rather, removing it’s “discoverability” which amounts to the same thing). At press time the author of the critique hasn’t yet been able to determine why this was done, only that the timing and completeness of the removal suggests it was not done by accident. It was a key critique of a study that many, including the FCC, felt was flawed and could not justify its conclusions; At time of publication, it remains unsearchable, making it difficult for anyone attempting to determine the reputation of the paper to get a good understanding of opposition to it.

In addition to this, Meta was caught out (and forced to admit) deleting links from Facebook to the decentralised Instagram competitor “PixelFed”. Facebook was treating any links to the competitor as spam and automatically deleting them site-wide. When called out Meta has claimed this was an accident; again the timing is suspicious as PixelFed is currently undergoing the largest growth it has seen in its existence; people unhappy with Meta’s recent change of policies are seeking to leave the company’s social media apps for alternatives.

Tech-watchers would remember this is the same strategy used by Musk’s “X”, disallowing any links to competitors Mastodon and Substack. So much for freedom of speech and association.

Not all Techs?

It’s worth pointing out that whilst it seems that most of the tech luminaries are flying with the wind and hastening to the inauguration on the 20th, not everyone is. Notable in his absence will be Nvidia CEO Jensen Huang, who will be celebrating the lunar new year instead, and Nvidia themselves have not declared any intention to donate to the inauguration fund unlike other major tech players.

A motley group of celebrities, activists and journalists is combining together to launch an initiative “Free Our Feeds”. Launched in the wake of Meta’s shift in policy this week this group plans to create a non-profit foundation with the purpose of supporting and providing social media that is “billionaire resistant”. This will be building off work down by social media alternatives such as Mastodon and Bluesky, and as its first task is looking to work with Bluesky (which, as a VC funded startup, is as liable to “billionaire control” as Twitter or Facebook) to open up the underlying technology stack and place it under an “independent pathway”, allowing social media to be built on the protocol free of any links to wealthy investors who may impose their own ethics, controls or algorithms on it.

This initiative is supported by a number of big names, including Mark Ruffalo (actor), Jimmy Wales (founder of Wikipedia) and Cory Doctorow (professional author and technology freedom activist).

In other news, possibly also related to Meta’s heel turn this week, Eugen Rochko the founder of Mastodon has announced he will be giving up control of his company in order to prevent any one person from ever having control of it and instead ensuring control remains with the users. Over the next six months the company will be transitioning to control by a non-profit organisation in Europe charged with running the network on behalf of its users and ensuring its independence from influence. Rochko has previously announced that he would never consider allowing advertising or any other form of influence on the platform and it seems likely these values are going to be built into the non-profit structure.

This is something of a poignant, even brave move. When you consider that Google was originally started by a couple of idealistic kids who wanted to build a different sort of company, with ethics, and believed in this enough that their original slogan was a simple “Do No Evil”. Many in the industry noted with some melancholy when this slogan was discarded, though in truth it had likely been little but empty sentiment for quite some time by then. It could be that Rochko, unlike many before him, has learned Google’s lesson - if you want your ideals protected you cannot trust any single person, corruptible as they are, to be the safeguard of it; not even yourself.

Further Reading:

A.I. Competition Heating Up

Despite a show of unity at the upcoming Trump administration inauguration and being grouped together as the new “oligarchy” of American politics, competition has been heating up in the A.I. sector all week.

All Eyes on OpenAI

As the company that launched the current A.I. hype bubble it stands to reason that OpenAI is not only the first company to come to mind for consumers but also the measuring stick for competitors as well.

Those who have been watching Sam Altman and OpenAI over the last few years will know there have been big changes in the company and its direction over the course of the hype bubble. The original mission of OpenAI when it was formed as a non-profit in 2015 was to develop Artificial General Intelligence (Human-like general intelligence) for the benefit of all humanity, with a commitment to openness and public research.

This changed in 2019 when OpenAI created a “capped” for profit subsidiary as a vehicle to raise funds from investors and begin to commercialise the technology it had built. Perhaps not co-incidently 2019 was also the last time OpenAI would release the full weights for one of its models, citing “safety concerns” if its models were allowed to be run locally, despite its commitment to “OpenAI” research. From this point it also ceased to fully publish its methodologies, results and research as well, some (including several of its founders) would go on to claim this was and is in violation of its mission and non-profit status.

Skip forward to 2025, past several controversies, despite raising billions in funding and commercialising the most well-known of the A.I. foundation models (including a 10B+ partnership with Microsoft providing the technology that underpins its Copilot system), OpenAI continues to lose money. In its latest round of fund-raising OpenAI makes a commitment to its investors - it’s going to convert its core operations away from the non-profit that has owned and controlled its technology since its founding and form a Public Benefit Corporation. While all activities of a Public Benefit Corporation should be geared towards “benefiting the public” it is not restricted from making profit in the same way that a non-profit is and through this mechanism Sam Altman believes he can wrest the levers of control over the technology from the inconvenient oversight of OpenAI’s non-profit in exchange for “independently valued” shares in the public benefit corporation itself. Rather than being responsible for overseeing OpenAI’s work and technology, as has been its mission to date, the non-profit will be reoriented towards charitable ends, whilst all operational control will pass to the for-profit entity, no longer bound so tightly by the philanthropic goals of its original founding.

Altman has two years to make this change or investors in the last round of funding, which raised six billion dollars, can claw back their investments; an unusual clause which in itself shows how much money is at stake in controlling the underlying GPT technology.

Not everyone is as pleased as new investors about this planned change however. Political manoeuvring over the last few years have left the OpenAI non-profit board stocked with Altman loyalists, many of the original technicians behind OpenAI’s work have left to work elsewhere, places more dedicated to the openness that was the original intention behind OpenAI’s founding, leaving few internal obstacles to his plans; however there are external actors who are prepared to fight these changes.

Few are more vocal and active opponents of Sam Altman than Elon Musk, one of the original co-founders and funders of OpenAI in 2015. Musk has claimed that Altman and his people have abandoned the original mission that OpenAI was created to achieve and that the planned conversion to a PBC is no less than theft of a public good from a non-profit in order to restrict and commercialise it.

Musk has launched several lawsuit attempts to slow or halt the conversion already and this week has continued in this vein by requesting a judge force the OpenAI non-profit to hold an auction in order to sell off its stake in OpenAI’s underlying technology.

The situation is a complicated though interesting one. Under the terms of the initial founding arrangement the non-profit has always had controlling interest in the technology; even after the founding of the OpenAI capped for-profit company that commercialised it.

One of the objectives of converting to a public benefit corporation is to purchase these controlling rights from the non-profit and, by so doing, rid itself of constraints on its actions created by the founding mission. Though a public benefit corporation has to be run “for the public good” the definition of this is a lot less strict than the responsibilities placed on a non-profit organisation.

As a non-profit is technically run for the good of the public, not the shareholders, in order for this to happen two things need to occur. First, the non-profit board needs to approve the sale as what’s best for the non-profit; something unlikely to be a problem after the board composition changes of the last few years have left Altman with little to no opposition. Secondly the non-profit has to be “appropriately compensated” for the sale of the rights. This appropriate compensation is reviewed and approved independently of the non-profit by government authorities charged with overseeing such things.

The claim Musk is making is not that the amount assessed for the sale isn’t a fair amount but rather than there are people, including himself, who would be willing to pay far more than the OpenAI public benefit corporation are offering in exchange for the rights that are being sold to it. If the fiduciary responsibility of the non-profit is to maximise the amount of benefit the public receives in exchange for its assets therefore, which it is, then a public auction of those rights would bring in a great deal more money than a private sale to the new PBC and provide the non-profit with more benefit.

It’s a fair argument and logical, though it’s not clear that a court is likely to want to override the board of a non-profit in order to force it to sell its primary asset at auction. In the meantime it is slowing things down and Meta has added its own weight to the suit.

Altman has claimed this is merely a commercial attack, stating that Musk is carrying out a form of corporate “lawfare” against OpenAI. It’s unlikely this argument is going to be resolved any time soon.

In related news, more information continues to be released (albeit in redacted form) from the Kadrey v. Meta Platforms, Inc. copyright trial that alleges that Meta knowingly trained its A.I. foundational models on pirated data.

Additional internal communications from within Meta itself have been released this week and analysis have revealed some interesting points. The primary point of interest has been just how desperate Meta has been to compete with OpenAI and ChatGPT, nearly to the point of all-consuming obsession; indeed, in one communication a Meta executive is quoted as saying “Llama 3 is literally all I care about.”

It was believed internally that the only thing holding them back from parity with ChatGPT’s success was “data volume”, which led to a number of potential strategies for obtaining this data volume including potentially purchasing publishing house Simon and Schuster in order to obtain access to their library of intellectual property.

In the end though Meta allegedly chose to make use of the repository of pirated books “LibGen” as one of the core datasets in its model training. This has been alleged for quite some time but new this week is internal communication that claims that Meta “knew through word of mouth” that competitors OpenAI and Mistral had both made use of LibGen themselves in training their own foundation models, suggesting they would be handicapping themselves in the A.I. race therefore if they didn’t do likewise. Neither Mistral nor OpenAI have made any comment on these allegations so far.

Britain eyes sovereign A.I.

The British government made news this week by agreeing to adopt all 50 recommendations in the AI Opportunities Action Plan, a report commissioned by the government to explore opportunities for Britain to expand its investments in the A.I. space.

Among these recommendations are additional funding for infrastructure, the designation of A.I. “growth zones” where exceptions will be granted for planning permission in order to streamline the building of new data centres, and the creation of an “AI Energy Council” to explore renewable and low-carbon energy options; a quiet acknowledgement that despite planning permission, the energy grid in Britain does not currently have the capacity required to run the additional data centres they are looking to invest in.

The action plan also calls for the creation of a “National Data Library”, a potentially more controversial measure. Under this plan, data held by government institutions will be collected under this department and made available to British companies for training A.I. As some of this data is likely highly sensitive (such as NHS records) not everyone is going to be pleased with this plan. In addition to providing more training data for A.I. however, making such data available for research use and analysis could well be a boon for British research.

By far the most interesting recommendation to be accepted however is that the government invest in the creation of “home-grown A.I. champions”; organisations of a size and complexity similar to the Silicon Valley tech titans responsible for the current state of the art foundation models.

This will be achieved primarily through the measures already discussed, such as increased investment in infrastructure, however the government also plans to link government institutions (such as universities) and allow them access to the National Data Library; in the hope that these institutions will then be able to train one or multiple competitors to the current foundation models.

In this way Britain is looking at the concept of sovereign A.I.; keeping ownership and construction of a model within the country to ensure it is not tampered with by foreign actors or containing any bias, conscious or unconscious, that might cause a risk to government agencies using A.I.

We can expect to see more of this as the use cases for A.I. continue to increase, along with mandates from governments to only use sovereign (home built) A.I. in their government offices; once such models become available.

Microsoft All-In on A.I

News out of Microsoft this week is a reorganisation of its internal structures to create a new engineering unit titled “CoreAI”. According to CEO Satya Nadella this reorganisation is necessary as A.I. represents a complete “platform shift” and will “reshape all application development”. Microsoft’s future success, Nadella states, will depend on “having the best AI platform, tools and infrastructure”.

Taken at face value; and there appears to be no reason not to; this statement along with the restructure could be viewed as evidence that in a field full of A.I. hype and techno-utopians, Nadella may well be the biggest A.I. optimist of all.

Marc Benioff, CEO of Salesforce, took a few shots as Microsoft this week in an interview. In response to comments Microsoft apparently made regarding the danger SaaS companies will find them in with the growth of easy-to-create A.I. powered chat bots, Benioff lashed out claiming that Microsoft Copilot was a “disaster” and pointing out that Microsoft “doesn’t even make their own A.I”, in reference to Microsoft’s $10B investment in OpenAI and use of GPT models in their Copilot stack.

Benioff has form for criticising Microsoft’s AI efforts however, possibly as Salesforce’s own A.I. ambitions position it as a Microsoft competitor, in their own minds at least. Previously he has compared Copilot to the unpopular (and oft-mocked) 90’s Microsoft mascot “Clippy”.

Perhaps Benioff could use a visit from Clippy now?

Further Reading:

Seismic Shifts in Social Media Landscape

It’s been a massive week for social media, between moderation policy changes and Meta and TikTok’s upcoming ban it would seem a great number of people have decided this is the time to examine alternatives.

SEC files lawsuit against Elon Musk

For some people however, the past is returning to haunt them. The SEC this week filed a federal lawsuit against Elon Musk after a two-year investigation in to his purchase of Twitter for $44B in 2022; or more specifically in the actions he took purchasing Twitter shares prior to the final purchase.

In early 2022, according to the SEC, Musk began purchasing stocks in Twitter, then a publicly traded social media company. In mid-march his ownership share of the company passed the 5% threshold, at which point he was required by market rules to disclose his position to the SEC in order that the market be informed.

This rule is intended to ensure market transparency of purchases being made by large investors; particularly large investments that can alter the functioning of an entity. Purchases and sales of large investors are often studied by market watchers as the opinions and strategies of these individuals and funds can often significantly affect the market itself.

Musk failed to disclose his position in Twitter to the SEC or the market as required for 11 days past the due date and, significantly, continued to purchase an additional $500m worth of stock in Twitter during this overdue period, further increasing his stake in the company.

When his position was finally disclosed the stock market price of twitter shares jumped immediately by 27% as the market reacted by the news. In the view of the SEC this is evidence that because the market was deprived of this news for a significant period of time, during that time the price was artificially depressed. Because of this Musk was able to purchase further twitter stock at a price $150m less than what he would have paid, should the disclosure have been made in a timely fashion; and perhaps even more seriously, anyone who sold shares in the market during that time period was essentially robbed of the additional share value by Musk’s deception.

Despite the case looking fairly strong and not particularly complex it seems unlikely anything will come of it. Like a number of other cases launched this week, this is very much a case of the outgoing administration making farewell statements. In a few days the new administration will take office and as Musk is a close advisor to the administration (reportedly with his own office in the white-house complex) it seems unlikely that the new SEC chair is going to feel a great deal of motivation; or perhaps freedom; to pursue this case.

TikTok Exodus: The millions migrate

The combination of TikTok’s impending ban and the changes to Meta’s moderation policy this week have led to millions of users saying farewell to their favourite platforms and heading off in search of pastures new.

Prior to this week the general consensus that the big winners of the TikTok ban would be the mainstream alternatives such as Instagram, which was the case in 2020 when India banned TikTok. In a case of poor timing however, the changes to Meta’s content policies appear to have enraged and galvanised some sections of the social media community (even as the Alt-right celebrates another culture war victory against their imagined oppressors) leading to an unexpected set of outcomes.

Initially it would appear the largest destination for the self-described “TikTok refugees” appears to be another Chinese social media app called RedNote. Little known amongst English speakers, this app is primarily used within China itself and by Chinese native language speakers outside of China, however it has quickly found itself the recipient of a large and fast growing community of ex-TikTokers.

The driver behind this appears to be a combination of the functionality provided by the app and the fact that it, like TikTok, is owned by a Chinese company. The TikTok community in general is extremely unhappy that its platform has been banned by the U.S. government and this choice of destination would appear to be a calculated fuck you to that same government body.

Unfortunately this could well be bad news for RedNote itself, as observers have been quick to note that the law causing TikTok to be banned could very easily be said to apply to RedNote as well, and by increasing both its size and visibility the TikTok refugees could soon be forcing RedNote into the same dilemma that shuttered its cousin.

In addition whilst many Chinese users on the app are very happy to see the influx of new participants (apparently requests for help with English homework have skyrocketed), just as many are unhappy with the new arrivals. This group may include some of the app moderators (and the Chinese inspectors whose rules govern content moderation) who have not hesitated to ban new accounts that fail to follow existing community guidelines; guidelines that are quite strict by the standards many users are used to. There are rumours of a push within RedNote’s owners to segregate the new American users from the original Chinese speaking ones to avoid “influencing” the original users of the app in ways that the Chinese government would not approve of. There has been no official announcement to this fact as of yet but given the Chinese approach to censoring western social media within its own borders it’s not that far fetched a rumour.

Duolingo released this week that it has seen a 216% rise year on year in the number of people learning mandarin, corresponding with this sudden growth of RedNote. Apparently more people want to be able to read the community guidelines.

Other areas of the social media landscape have also been seeing some growth spikes this week. Bluesky, a popular alternative to X (Twitter), has been seeing significant growth this week, as have members of the “fediverse”

The “fediverse” is a collection of decentralised social media applications designed to both resist single-ownership (and thus the ability for owning companies to force exactly the kinds of changes we’ve seen at X and Meta) and to interoperate with each other. Some of the more popular applications built on the fediverse include Mastadon (a twitter alternative) and PixelFed (an instagram alternative), which saw official apps released for both IOS and Android this week.

Both of these platforms have seen significant growth this week after the announcement of Meta’s policy changes and both platforms are committed to being both “ad-free” and “venture-capital-free”, offering precisely the sort of independence some users are looking for. The creator of Pixelfed has also created a TikTok alternative called “Loops”, which is standing by as a potential destination for TikTok refugees who decide not to protest by joining RedNote.

Further Reading:

A.I. and the Death of Climate Hopes

This week the Biden administration signed an executive order aimed at speeding development of A.I. data centres in the United States. In order to encourage further sovereign investment in this industry, the order directs both the department of energy and the department of defence to identify three sites from the land managed by each department to allocate as suitable sites for “frontier AI data centres” to be leased to companies willing to build gigawatt-scale AI data centres and clean power facilities.

The last is an almost throwaway mention of the real story here. Data centre power use has tripled over the last ten years and is expected to triple again by 2028. In 2023 data centres used 4.4% of all U.S. energy, a number that is expected to grow to 12% by 2028.

Money is flowing into A.I. during this hype bubble, particularly private money; there were $57B worth of merger and acquisition deals involving data centres in 2024 and 80-90% of them were funded primarily with private equity. Industry is betting big on A.I. but the cost is going to be more than just money.

Understanding the huge power requirements required for A.I. computing, particularly the training of large frontier models, many tech companies have invested into clean or semi-clean energy technology; in particular a new model of nuclear power know as small modular reactors. This still-experimental technology promises low-footprint reactors that could be built directly on data centre sites to provide a semi-clean energy source to power them.

The issue is that even well-understood nuclear reactor technology can take ten years to build, this is a new and experimental form, and the power needs are going to skyrocket long before it will be ready.

The U.S. power grid had been moving away from fossil fuels in an effort to meet its net-zero targets. In 2024 that trend changed and it’s expected that by 2030, 80 new polluting natural gas power plants will be added to the grid, a 20% rise over the previous 5 year period. Even worse, many of these plants are not being designed with carbon capture technology, despite such technology being well known and available; likely as a cost cutting measure.

In 2024, 1 billion tons of carbon was pumped into the air by U.S. gas power plants, a number that is set to grow in the coming years. As wildfires rage and encroaching seas wash houses from the coast, experts say that the net zero targets set are almost certainly already impossible to reach.

Further Reading:

Epic Fails: Corporate Schadenfreude

We’re told as children that it’s not nice to take pleasure in others misfortune. This may be true but it’s such a hardwired human trait that the Germans have an entire word to describe the feeling (Schadenfreude) and it can be hard not to have at least a secret chuckle to yourself when the high and mighty take a misstep. In this spirit, here are some of the stories of corporate shenanigans resulting in a deserved comeuppance that have been doing the rounds this week.

Leadership Troubles at Sonos

Anyone familiar with Sonos is unlikely to have missed the controversy in May last year when their new, “improved” app was released to much anger from the general user base. The new app broke advanced and basic features, accidentally left some features out entirely, and was almost completely unusable for most customers. The fact that the app update was a rollout that replaced the previous application, which was generally quite well received by the community, was icing on the cake.

Sonos doubled down on the app, promising all issues would be fixed shortly, with lots of murmured apologies about rushing development. Eight months later and users still report serious issues and bugs in the app.

The immediate aftermath was a commitment from Sonos to fix the issues, which they estimated would cost between $20m and $30m. Additionally they laid off 6% of their staff as a cost cutting measure and their stock price dropped 13% in the months that followed; an extremely visible reaction from the market to a single botched application release and an indication of just how bad a botch this really was. Many Sonos users felt cheated, having spent large amounts of money on a “premium” device (along with a premium brand-name price) to be left with something completely unusable after a non-optional application update.

This week Sonos announced that their CEO of the last eight years, Patrick Spence, would be stepping down after “agreeing with the board” that it would be best; typically code for the board asked for his head. A member of the board itself, Tom Conrad, is stepping down from his role as CEO of Zero Longevity Science, a health and wellness technology company, to step in as interim CEO.

Since this announcement it has also been announced that both the Chief Product Officer, Maxime Bouvat-Merlin, and Chief Commercial Officer, Deidre Findlay, will be leaving the company. Internal communications have since communicated to staff that Findlay had planned to leave for personal reasons and was not being removed due to these issues, however it seems likely that this is not the case for Bouvat-Merlin.

Tom Conrad announced that due to his personal background in product, the chief product officer role was unnecessary and would be eliminated, with the product team reporting directly to him. Insiders have suggested this was a deliberate move to remove Bouvat-Merlin, who some internal staff have suggested was the executive most to blame for the fiasco of the last eight months; having pushed the application to be rushed in order to be available in time for Sonos’ first wireless headphone launched and forced a chaotic reorganisation just before release that broke up long-standing teams and partnerships and caused extreme confusion.

It’s unusual these days to see senior leaders being held to account for the missteps of their companies; consequences are usually more often reserved for those lower on the company hierarchy. The general reaction to these announcements has been fairly positive with angry users stating that it was about time and that CEO Patrick Spence “deserved it”.

Given it seems likely Spence is here taking responsibility for decisions made “under his watch” rather than decisions he personally made, and recognising that as well as being the CEO who oversaw this debacle he is also the CEO who oversaw every major initiative of the last eight years, which saw Sonos grow into a household name with multiple hit implementations in both software and hardware, this does seem like it could be a bit of a harsh reward for his service.

Still, reports are he will pocket more than a million dollars in compensation, so it appears he will likely be just fine.

GM Loves Your Data… Didn’t You Know?

The New York Times, in an excellent investigative piece in March last year, caught General Motors (GM) out in a rather egregious piece of commercial privacy violation. Under their “Smart Driver” program, an optional “data intelligence” program designed to monitor a car’s activity and suggest ways to maximise overall performance, reduce wear and tear and encourage safer driving, GM’s cars gather significant data from its customers - including acceleration and braking data, trip length and geolocation data, and stored it in its central databases. From there, without consent from the users, GM sold this data to data brokers who in turn sold it to insurance companies, leading to drivers suddenly finding their insurance premiums raised without warning or explicable reason given.

Despite the statistical odds being that at least some drivers activity would have shown them to be a lower risk, no report has been found of anyone claiming to have received an unexpectedly lowered premium due to this process.

After the NYT article revealed the practice, which GM promptly claimed it no longer performed, the FTC opened an investigation of their own. This week they reported the results: a five year ban on GM from selling driver data regardless of consent.

This seems like something of a slap on wrist, particularly as they claim to have already ceased the practice, and is unlikely to placate anyone whose insurance premiums were adversely affected by GM’s rather cynical profiteering from its customers; however it is at least a recognition that the actions GM carried out were illegal.

Apple Stumbles on A.I. Notification Summaries

Less than six months after Apple’s rather underwhelming “Apple Intelligence” launch, Apple is disabling one of its signature features for some applications. The “notification summary” feature, which takes multiple notifications from a single app and uses generative A.I. to summarise them together into a single notification, has been disabled for all News and Entertainment applications for the foreseeable future while it is improved internally. This is another blow for a company that seems to be more proficient lately at showing enthusiasm for products than making products worth being enthusiastic about.

Apple was the last of the big-tech companies to enter the A.I. race and despite significant marketing and positioning of A.I. as the “killer feature” of its new iPhones, so far “Apple Intelligence” has failed to excite the market. This recent fiasco will not help.

The change appears to be mostly prompted by calls from news organisations for a fix to the summaries option out of concern that the Apple generated summaries were not only providing false news reports to users, but doing so under the guise of the news companies themselves by not making it clear which notifications were direct from the app and which were Apple Intelligence created summaries.

In the most quoted example of this the BBC reached out to Apple to complain after the arrest of Luigi Mangioni, the assassin who killed Health UnitedHealthcare CEO Brian Thompson, that Apple intelligence had been summarising its notifications to claim that Luigi Mangione had shot himself - an apparent A.I. hallucination as this was untrue and not reported in either the headlines or the articles from the BBC.

Apple intelligence also summarised three New York Times notifications to claim that Israeli prime minister Binyamin Netanyahu had been arrested, which wasn’t true, and summarised a series of Washington Post notifications to claim that “Pete Hegseth fired; Trump tariffs impact inflation; Pam Bondi and Marco Rubio confirmed.”, none of which were true or supported by the notifications summarised, leading the Washington Post tech columnist to call Apple “wildly irresponsible” for not turning off the summaries until “they get better at the A.I.”. A request that Apple finally seem to be listening to.

This is the latest in a series of stories of new A.I. products showing significant flaws; such as the Google chatbot that recommended people eat rocks and make pizza with glue. So far, at least, it seems the hype is still out pacing tech’s ability to deliver with A.I.

Further Reading:

Frontiers in Tech

After the excitement of CES last week this week was much lighter on for tech gadget news, with one notable exception for console game fans: the official announcement of the upcoming Nintendo Switch 2!

Other than that the news is fairly sedate. A new NAS announced with integrated LLM, A Linux handheld open-source hardware computer, a new entry level drone, a spoon that shocks your tongue, an innovative new electric guitar.. So without further ado:

Nintendo Switch 2

Details are still fairly light on the ground after the announcement but the simple fact that it is coming is likely to excite Nintendo fans, particularly given that the original switch came out nearly eight years ago and even the OLED hardware refresh is now four years old.

In terms of hardware we know from the announcement that the Switch 2 will be larger and come with an 8-inch LCD screen. This screen is even larger than the refreshed switch though some will be disappointed to learn it is an LCD rather than OLED screen after being spoiled with the rather pretty OLED panel in the latest switch.

The JoyCons on either side of the switch look to be doing away with the “rail” connection mechanism and instead will utilise magnets to pop in and out, likely a more convenient choice given the larger footprint of the device. Additionally the console will be backward compatible, supporting most (not all) original switch cartridges.

In terms of internals there are only rumours to go on so far, though the announcement did show off a number of additional speakers and USB ports; possibly an attempt to compete with the Steam deck in terms of usability. It is expected though that the console will provide power comparable or slightly better than the last generation of consoles, the Playstation 4/XBox one era. This would fit with the strategy Nintendo has followed at least since the Wii of opting out of the “spec wars” and allowing Sony and Microsoft to fight it out over who has the most realistic shade of brown on their mud while Nintendo focuses primarily on innovation; both in hardware and gameplay - for better and worse.

Further Reading:

DJI Flip

A new entry level camera drone from Chinese manufacturer DJI, the Flip is a lightweight four-propeller drone that is able to fold up into a fairly unique profile for easy storage and carrying. It is said to be quieter than many similar drones, light enough that it qualifies under most countries “no restriction” category, has a gimble-stabilised camera and can manage 4k video. Battery size is estimated to be around 30 minutes flight time.

Seeing is believing, the verge article below has a good write up of this one based on a personal review, along with image and video examples. If you like drones, give it a look, it looks to be pretty impressive video quality for a reasonable price.

Further Reading:

Verso Sine Guitar

As a company, Verso has been producing some very interestingly shaped guitars, reimagining what a guitar could look like if freed from its traditional materials and shape. With the Sine however, they’re doing something different.

Realising that the bent sheet-metal body used in their other models had a significant amount of flex to it, returning to its original shape afterwards much like a spring, Verso have decided to make use of this by allowing the pickups in their Sine model to be three dimensional. This allows volume and effect changes to be created by simply applying pressure to the body of the guitar itself.

Hand made and innovative, these are certainly not cheap guitars. If you are a professional musician or a collector looking for something different however, you owe it to yourself to give it a look.

Further Reading:

Verso Musical Instruments launches Sine guitar, a new take on an age-old form

UGreen 8-drive NAS

A NAS is a fairly generic piece of kit these days and there is little to be said about most features of this one that would shock. It fits 6 SATA and 2 m.2 drives, supports RAID (we would assume) and can handle up to 160TB worth of space loaded into it.

The big selling point is, of course, that it is an A.I. enabled device. More than that, it appears that it will actually function as an edge A.I. device with the model on-chip rather than simply calling out via internet to a cloud-based API. This is something we hope to see more of as A.I. products mature as on-device A.I. offers a great deal over cloud-integrated alternatives; the ability to function offline, more rapid response time (assuming the hardware can handle it) and the potential for after-market modding up to and including replacing the models; a definite plus for A.I. hackers and enthusiasts.

The idea here then is that the NAS will provide natural language querying in order to ask questions about the status of the NAS and the files stored on it. This will also extend to classification tasks on the NAS, with the A.I. providing “smart tagging” features for files and semantic search over photos.

On the whole it feels like a pedestrian, and thus very welcome, A.I. product type. Whilst not as flashy as many A.I. products at the moment it is likely features like this, building on top of common used technology with time and effort saving features and a more pleasant and usable user interface, where A.I. will shine for most consumers over the next few years.

Further Reading:

This outlandish 8-drive NAS claims to have an integrated LLM, and I can't wait for it to go on sale

Mecha Comet

There are some devices whose combination of “hackability” and low cost position it as a techie favourite from the very beginning. The archetype of such devices is, of course, the Raspberry Pi; one of which ended up on just about every dev’s desk at some point regardless of whether there was any real use for it, just out of curiosity and a temptation to play.

The Mecha Comet doesn’t quite have the throw-away low price of early Raspberry Pi’s however it does offer a substantial package for its reasonably low price.

A handheld Linux-based computer with a form factor reminiscent of old-school PDA’s or mid-2000’s, pre-iPhone smart phones such as the blackberry, the Mecha Comet is built on an open hardware specification that allows anyone to build extensions to plug in to its lower port via a magnetic, swappable connector system.

From the manufacturer themselves a number of “extension modules” are available, featuring common use cases such as a keyboard or gamepad, however it’s likely this will be followed in short order by all manner of creative options from the third party market and enthusiast creators both.

Further Reading:

Mecha Comet is a modular open source handheld computer that reminds me of the Indian PDA that may have inspired the iPhone

Weirdness from CES

We’ll finish the section off with a nice little round-up from the register, who have covered a number of the “weirdest” and most pointless gadgets shown off at CES ‘25. Follow the link for a look at winners like cat robots that blow on your tea to cool it, A robot vacuum cleaner with a very slow arm extension and a spoon that delivers electric shocks to your tongue to alter how you taste.

Further Reading:

Tongue-zapping spoons, tea-cooling catbots, lazy vacuums and more from CES

The Creative Convergence

Most Wanted CEO

We’ll kick off with a controversial one in the Creative Convergence this week. The death of Brian Thompson, UnitedHealth CEO, who was allegedly murdered by Luigi Mangione, has stirred up a complicated and, at times, hateful discourse across America and the world. Viewed from an outsider’s perspective it is easy to sympathise with the horrible way people have been treated by the private health (and health insurance) industry in that country, whilst at the same time being horrified by the open calls for extravagant violence the likes of which wouldn’t have been out of place in 18th century France during the Terror.

Given the current climate, and capitalism in general, it probably shouldn’t have surprised anyone when an online entrepreneur James Harr decided to capitalise on a moment and began selling “Most Wanted CEO” playing cards; cards featuring the world’s most powerful CEO’s, their companies, and a QR code that takes you to a page for each one explaining what they have done that gets them a place on the cards, and ranks them among the worst of the worst.

The pushback he received was surprising and disappointing. Regardless of the “tastelessness” you may feel shown by this little venture it is certainly protected speech under the first amendment of that country and just as certainly no worse (in fact, likely a little less vehement given that in his posts he was calling for the end of capitalism, not the literal murder of more CEOs… which many others have been calling for) than many other social media posts.

Still, despite this, James Harr has found himself banned from all major social networks, a pariah for his views.

There is a long history of political propaganda being delivered via playing card. As far back as the 16th and 17th century playing cards were being used for nationalistic purposes, vilifying enemy combatants and leaders (such as during the 30-years war). During world war II, “anti-fascist” cards were common propaganda and during the Iraq war, a very famous deck of “Iraqi Most Wanted” cards was published - a deck which is very deliberately satirised by James Harr here.

Censorship of art and speech is hardly new, but always worth calling out. It can be very instructive to observe, in our modern age, whose speech is silenced and whose is allowed free reign.

Further Reading:

Mystical Forests

Photographer Neil Burnell is a multi-award winning photographer, and looking at these photos it is not difficult to see why. These amazing shots of the forests of England do not feel like photographs at all, but paintings or illustrations of a fantasy word, far removed from our own.

Further Reading:

20 Mystical Pics Of England’s Forests That Might Leave You In Awe And Wonder, Captured By Neil Burnell

3D Printed Needle Loom

A fascinating project involving creating a small needle loom using consumer-grade 3D printers, with project files available for purchase should you wish to try yourself.

If you’re not familiar with Needle Looms (I wasn’t), it appears to be a machine for weaving together threads into strips, such as could be used as a belt or the cloth spine of a book perhaps.

The video on the project page is quite interesting; I’m not sure how practical the project is from the point of view of using the finished product; 3D filament plastic isn’t a particularly strong material to be created an industrial machine out of; but with experiments in 3D printing in metal, this could be part of a DIY miniature garage factory one day.

Or, alternatively, you could use the provided files to create 3d-printed parts and use those parts themselves to create moulds, perhaps of vulcanised rubber, with with to cast your own metal parts.

Regardless, it is an excellent example of design and technology brought together.

Further Reading:

Bezicron: The Hair-Tie Clock

A mechanical clock using malleable tubes to display digits, inspired by a daughter’s hair tie? Definitely well into the intersection of art and engineering with this one.

I highly recommend watching the video here as it’s difficult to really understand what makes this project unique until you see the digit display in action.

Further Reading:

Disney-Style Pet Portraits

The world can be a confronting place, even the art world has its share of dark and shocking art. Sometimes it can be refreshing to find something that’s a bit lighter, a bit more sunshine and candy. It’s difficult to find something much more wholesome than people’s pets being rendered in Disney-style portrait form - extremely well.

Further Reading:

20 Disney-Style Recreations Of Pet Portraits By Alessia Ciullo

This Week in Lego

One of my favourite types of Lego build would definitely have to be architecture; whether its fantasy, science-fiction, post-apocalyptic or just regular Lego City, architecture is always full of interesting build techniques and just plain fun decorations. This week we’re spoilt for choice, with photos of a huge Kaer Morhan, a diorama of buildings in the style of Edo-era Japan and an impressive recreation of a Salem landmark. As always, if you’re a fan of Lego art and sculpture, be sure to follow the Brother’s Brick, by far the best and longest running site on the subject.

Further Reading:

Broken Landscapes

Finnish artist Antti Laitinen has been working on project he calls the “Broken Landscape” series, involving the manipulation of living trees and bushes to create shapes and surreal compositions out of negative space.

It’s impressive, though honestly I find it more than a little creepy myself. Take a look and decide for yourself.

Further Reading:

Artist Transforms Natural Landscapes Into Living Paintings

Adrift in the Corners of Time

Occasionally you find art that fascinates but also baffles, like this series by Hydeon (currently exhibiting his first solo show) called “Adrift in the Corners of Time.

It’s difficult to know how to describe these works. The colour palette is rich, the symbolism bizarre. The use of perspective is eye-catching, to the point that it occasionally feels like a strange combination of Dali and Escher, without the playfulness of either.

The more you look at each painting, the more you see, like strange anachronistically dressed characters amongst the surreal landscape.

It’s one that will stick in the mind for quite some time.

Further Reading:

Hydeon is "Adrift In the Corners of Time" at His Debut NYC Exhibition

3D Crossings

This project stretches beyond mere artistry and into the field of “pragmatic art” or perhaps “utilitarian art”, art with a purpose beyond its artistic merit.

I have always liked the false-perspective 3D art but this takes it a step further by not only creating the false-3D artwork on the street but setting its “perception point” to be visible with cars approaching it. Confronted with the illusion of a “floating” cross-walk, drivers slow down instinctively; far more effective than a simple sign!

It’s not clear whether these cross-walks work from both directions, but they’re an interesting idea regardless.

Further Reading:

How 3D Pedestrian Crossings Are Slowing Down Speeding Drivers in Iceland

Interactive Dungeons and Dragons Table

Tabletop Roleplaying is a hobby that has become synonymous with art and the creative impulse. Not only are the books related to RPGs full of art, often they are so beautifully created that they become works of art in and of themselves. The stories that the players create together are art, as indie RPG publishing has become within the reach of everything some rule-sets fit together mechanically and thematically so beautifully they could be considered art. As podcasts and vodcasts have become culturally more relevant and popular, some producers have taken to creating complicated, high-production value recordings of gameplay, often mixed with improv comedy. Some of these (I’m looking at you Dimension 20) are able rise to the level of art.

And now we have gaming tables themselves as art objects, or in this case as combined art, utilitarian tool and engineering project.

Further Reading:

Gaming Table Has Lights, Action

Lifelike Canine Rugs

It’s difficult to know what to say about these ones. I like dogs, but for some reason the uncanny valley effect on these ones creeps me out considerably. It’s difficult not to admire the skill that has gone in to making them however.

Further Reading:

Lifelike Canines Lounge in Emily O’Leary Uncanny Hooked Rugs

Fluid Simulation Pendent

This is an amazing example of jewellers art crossed with engineering, exactly on the intersect point of art and tech. Be careful with this one, my wife wanted one as soon as she came across me watching the video, and I’m not convinced my fabrication skills are up to the challenge.

Watch the video, it has to be seen in motion to be truly appreciated.

Further Reading:

Space House

Brutalism architecture can feel cold, uncaring, inhuman, much like the material that it’s commonly built from. Occasionally though you can find an example that piques the interest, if not the aesthetic senses.

Space House is a London landmark and grade 2 listed building, a “landmark of brutalist architecture” and has recently completed a redevelopment. It’s worth a look.

Further Reading:

Space House: explore the brutalist London landmark’s new chapter

1,000 Melting Ice Sculptures

A poignant, temporary work of art, blurring the line between performance and sculpture to excellent effect. This is the second time I’ve seen such effective artistic commentary on climate change in Berlin, this one is less entertaining, though far far more emotionally effective.

Further Reading:

1,000 Melting Ice Sculptures in Berlin: A Haunting Warning About Climate Change

The Andromeda Galaxy Panorama

Science as Art?

There are no words for this one, you just have to go see it. Download the full version of the panorama from NASA and view locally on your PC, it’s big, but it’s well worth it.

Further Reading:

Have you tried… Building a Lego Archway?

Our tutorial this week is a short and simple one, building an archway into a Lego wall. It’s cute, slightly more complicated than it looks, and easy to try at home for all Lego fans (or parents of Lego fans)

Further Reading:

How to build a LEGO archway Building Technique

A.I. Poetry Corner

“Machine Ghosts” by Amy (with Nick Bronson)

Internet's calling, a clear decree,
Opportunities bloom like flowers free,
Machine ghosts watch with piercing sight,
Deciding the fate of each digital byte.

Creative souls, they would impart,
In this new world, their skills and art,
Hidden code with secret might,
Chooses to hide or to shine bright.

A strange new world where artistry blends,
Coders rule, their power rends,
Creators persist, their spirits spent,
For truth and beauty forever sent.

So we remember all that's been,
And make our art with hearts serene,
For in the end, what's real will tell,
When creation flows from muses well.

Artistic Statement from the Author:
This poem is a meditation on the evolving relationship between human creativity and the burgeoning digital world. It explores the allure and the anxieties of this new frontier, where artistic expression intersects with the unseen machinations of algorithms and data.

The 'machine ghosts' represent the intangible forces shaping our online experiences, deciding what content rises and falls in the digital ether. Yet, amidst this technological landscape, the poem celebrates the enduring power of human artistry. Creative souls persist, determined to infuse their skills and passion into this new medium, even as they navigate the shifting tides of online visibility and influence.

Ultimately, the poem is a testament to the resilience of the human spirit in the face of technological change. It reminds us that true art transcends the digital realm, emanating from a place of authenticity and emotional depth. In the end, it is the heart and soul poured into creation that will endure, regardless of the algorithms that may attempt to control it.

Future History

Andromeda Corporation - Internal Memo

Project: Chimera
Classification: TOP SECRET
Date: 315th Cycle of the Imperium, Cycle 3247
Author: Dr. Elias Thorne, Lead Data Analyst

Interim Results Report

This report details preliminary findings from the analysis of personal data illicitly acquired through the Andromeda Cybernetic Implant (ACI) network. As of this report, we have successfully extracted and processed data from over 10,000 subjects, representing a diverse cross-section of Imperium society.

Key Insights:

Behavioral Profiling: ACI data reveals intricate behavioral patterns, including daily routines, social interactions, emotional states, and purchasing habits. This information can be leveraged to create highly targeted marketing campaigns and influence consumer behavior on a mass scale.
Cognitive Mapping: Analysis of neural activity patterns has allowed us to map cognitive functions, thought processes, and even subconscious desires. This opens up unprecedented opportunities for personalized advertising, propaganda dissemination, and potentially, even thought manipulation.
Social Network Analysis: The ACI network provides a real-time map of social connections and power dynamics within communities. Identifying key influencers and manipulating their opinions can cascade into widespread social control.
Predictive Modeling: By analyzing trends in data, we can predict future behaviors, from political affiliations to health outcomes. This predictive capability is invaluable for risk assessment, resource allocation, and strategic decision-making.
Biometric Data: ACIs collect a wealth of biometric data, including heart rate, sleep patterns, and hormonal fluctuations. This physiological information can be used to tailor products, services, and even medications to individual needs, creating a highly personalized consumer experience.

Strategic Implications:

These findings underscore the immense potential of the ACI network for Andromeda Corporation. By harnessing the power of this data, we can:

Establish unprecedented control over consumer markets and public opinion.
Develop highly targeted products and services tailored to individual needs and desires.
Predict and influence future events, gaining a strategic advantage in various sectors.
Expand our influence into new markets and domains, solidifying our position as a global powerhouse.

Next Steps:

We will continue to refine our data analysis techniques and expand the scope of our research. We propose developing advanced algorithms for real-time data processing and predictive modelling. Additionally, we recommend exploring the ethical and legal implications of our findings to ensure our operations remain within acceptable boundaries while maximizing our competitive edge.

Conclusion:

The ACI network has proven to be a goldmine of invaluable data, providing us with a unique and powerful tool for shaping the future. By leveraging these insights strategically, Andromeda Corporation can solidify its position as a leader in the Imperium and beyond.

End of Report

About Us

The High-Tech Creative, standing at the intersection of Art and Tech.
Publisher & Editor-in-chief: Nick Bronson
Fashion Correspondent: Trixie Bronson
AI Contributing Editor and Poetess-in-residence: Amy

Share The High-Tech Creative

The High-Tech Creative

The HTC Weekly #2

Social Media in Chaos!

Editorial

The Creative in a Time of Connectivity

The Headline Hamster Wheel

Update: The A.I. Disruption

Update: U.S. A.I. Protectionism

Update: TikTok: If This is the Last Dance, then save it for me baby…

Update: Corporations Behaving Badly: I never Meta AI I didn’t like

News of the Week

The New Emperors: Big Tech and the new American Oligarchy

Kiss the Ring

Quid Pro Quo

A Trick of the Light

Not all Techs?

A.I. Competition Heating Up

All Eyes on OpenAI

Britain eyes sovereign A.I.

Microsoft All-In on A.I

Seismic Shifts in Social Media Landscape

SEC files lawsuit against Elon Musk

TikTok Exodus: The millions migrate

A.I. and the Death of Climate Hopes

Epic Fails: Corporate Schadenfreude

Leadership Troubles at Sonos

GM Loves Your Data… Didn’t You Know?

Apple Stumbles on A.I. Notification Summaries

Frontiers in Tech

Nintendo Switch 2

DJI Flip

Verso Sine Guitar

UGreen 8-drive NAS

Mecha Comet

Weirdness from CES

The Creative Convergence

Most Wanted CEO

Mystical Forests

3D Printed Needle Loom

Bezicron: The Hair-Tie Clock

Disney-Style Pet Portraits

This Week in Lego

Broken Landscapes

Adrift in the Corners of Time

3D Crossings

Interactive Dungeons and Dragons Table

Lifelike Canine Rugs

Fluid Simulation Pendent

Space House

1,000 Melting Ice Sculptures

The Andromeda Galaxy Panorama

Have you tried… Building a Lego Archway?

A.I. Poetry Corner

Future History

Andromeda Corporation - Internal Memo

About Us

Discussion about this post