New World Same Humans

Welcome to this update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

For this week’s instalment, I’m doing something different.

A few days ago I recorded a video update to share some thoughts on OpenAI’s new GPT-4o and the state of AI.

It went first to Exponentialist subscribers. But I want to share it with you all, too.

In the video I get into:

* Why GPT-4o is OpenAI’s play for billions of users, and for a virtual companion that weaves itself through the fabric of everyday life

* Where we are inside the amazing AI moment we’re living through, and what’s coming next, including a path to AGI

* How this all connects to the Great Enweirdening of the economy that I believe is coming

There’s so much happening with AI right now; I hope this provides some useful framing. And if it proves popular, I’ll do more video updates in future.

By the way, there’s still time to grab a six day trial to The Exponentialist for just $1.

Thanks for watching, and be well,

David.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to this update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

This week brings news from Boston Dynamics and the Chinese Academy of Sciences. The message common to both stories? The humanoid robots are coming.

Meanwhile, the internet reacts to Apple’s new Vision Pro headset.

And the FCC take action against a California company that used AI to create fake phone calls from President Biden.

Let’s go!

🤖 Robots are go

This week, yet further signs that the robots will soon walk among us. I mean, all of us.

The Boston Dynamics humanoid, Atlas, has been a regular in this newsletter over the years. Recently it has been overshadowed by competitors, including the Digit humanoid by Agility Robotics and Tesla’s Optimus.

But this week Boston Dynamics released a video that shows Atlas picking up automotive struts and placing them in a flow cart.

The team say Atlas is using onboard sensors and object recognition to perform the task. The footage is short. But it marks a significant advance for Atlas, because previous videos have shown the robot doing elaborate dances rather than useful work, and those dances have been pre-programmed rather than autonomous.

Meanwhile, in Beijing a research team at the Institute of Automation in the Chinese Academy of Sciences this week debuted their Q Family of humanoid robots.

The research team have reportedly built a ‘big factory’ for the design and manufacture of Q Family humanoids.

Back in New Week #124 we saw how the CCP has ordered ‘domestic mass production’ of humanoids’ to fuel economic growth. Remember, this is the underlying demographic reality that has China dashing towards robots.

⚡ NWSH Take: In last month’s Lookout to 2024 I said this would be the year of the humanoid. We closed out 2023 with the announcement that the Digit humanoid had started a trial inside US Amazon fulfilment centres. Days after I published the Lookout, BMW announced a trial of Digit in its California manufacturing plant. Now, the Boston Dynamics team are clearly eyeing commercial applications, too. Their Atlas robot has so far remained a research project; the question they’ll have to answer if they want to change that is whether Atlas can match Digit and Tesla’s Optimus for autonomous capability. // The graph above tells the underlying socio-economic story here. Both the CCP and innovators in the Global North know that working age populations are falling. If economic growth isn’t to become a distant memory, we need new armies of autonomous workers. AI applications can handle some of our knowledge work. But we’ll need humanoids to do some of the physical work that currently only people can do. The CCP see this as an existential imperative; they know they must maintain GDP growth. For innovators in the US and beyond, it’s an epic opportunity.

👀 Having visions

No one could have missed the launch of the Apple Vision Pro a few days ago.

Years from now, this instantly iconic magazine cover will no doubt spark intense nostalgia for the simpler times that were 2024:

It took about ten minutes for someone to try out their new Vision Pro while using Full Self Drive in their Tesla:

This was later revealed to be (surprise!) a skit for YouTube. Still, it delivered useful findings; the man in the picture, Dante Lentini, says the Vision Pro doesn’t really work inside a moving car because it can’t properly display visuals over a fast-moving landscape.

⚡ NWSH Take: After the frenetic metaverse hype of 2021, many will shrug at the launch of the Vision Pro. But something real, and powerful, is happening here. The internet is going to become part of the world around us. In the end, this is about the deep merging of information and physical reality, of bits and atoms, that I wrote about in the essay Intelligence in the World. // We’re going to see the emergence of a unified digital-physical field: a blended domain of bits and atoms that is a new, and in some sense final, innovation platform, because it brings together everything we do online with everything we do in the real world. // Apple’s new product — whether it proves a hit or not — is just another signal of this underling process. I’ll get my hands on one ASAP and report back. But Apple, here, are clearly aiming at high-end and industry users; they’re going to have to maker a cheaper product if they want mainstream impact.

☎️ Good call

Also this week, a glimpse of what lies ahead when it comes to this year’s US presidential election.

The FCC this week banned AI-voiced robocalls after an AI Joe Biden ‘called’ over 25,000 voters in late January and told them not to vote in the then-upcoming presidential primary elections.

The calls have been traced back to a Texas-based company called Life Corporation, owned by an entrepreneur with a long history in automated calling for political campaigns. Researchers believe Life Corporation used software from UK-based AI voice startup ElevenLabs, which I’ve written about here several times before, to deepfake Biden’s voice.

ElevenLabs just raised an $80 million series B funding round, led by VC firm Andreessen Horowitz, that valued the company at $1.1 billion.

⚡ NWSH Take: In the Lookout to 2024 I said we should expect politics to collide with the exponential age this year. The impact of AI deepfakes on November’s US presidential election will be at the heart of that story. Okay, the FCC has banned AI calls. But deepfake audio and video is surely going to be rife on Facebook, Elon Musk’s X, and TikTok. // Our liberal democracies were built in the age of one-to-many mass broadcast; those broadcasts were gatekept by social elites that felt a sense of duty towards the broader socio-political system in which they were operating. It wasn’t perfect, but it muddled along. Now, we’ve built previously unimagined technologies of image and sound manipulation. We’ve slain the gatekeepers, and told ourselves that this was an empowering move. The upshot? We're about to find out how liberal democracies work under those conditions.

🗓️ Also this week

👶 Researchers trained a large language model using only inputs from a headcam attached to a toddler. A data science team at New York University strapped a camera to a toddler for 18 months. They say their AI model learned a ‘substantial number of words and concepts’ from exposure to just one percent of the child's total waking hours between the ages of six months and two years. The team say this indicates that it is possible to train an LLM on far less data than previously believed.

🏭 Sam Altman says the world ‘needs more AI infrastructure’ and that OpenAI will help to build it. Altman is reportedly seeking trillions of dollars to build new semiconductor design and manufacture capability. Access to chips and the compute they supply is crucial for OpenAI if they are to train GPT-5 and other large AI models.

💸 Disney says it will invest $1.5 billion in Epic Games, the makers of Fortnite. The media giant say they’ll work with Epic to create a new ‘entertainment universe’ featuring characters from Pixar movies, Star Wars, and more.

🦹‍♂️ The US National Security Agency say an advanced group of Chinese hackers have been active across US infrastructure for at least five years. The Volt Typhoon hacking group is said to have infiltrated computer systems across aviation, rail, highway, and water infrastructure.

🔋 Europe’s deepest mine is to be converted into a gravity battery. The Pyhäsalmi Mine in Finland is 1,444 meters deep. Its copper and zinc deposits have run out. Scottish energy tech firm Gravitricity say they will now convert the mine into a gravity battery, in which energy is created stored via elevated heavy weights and released when those weights are dropped.

💥 Scientists at CERN want to build a massive new particle collider. The new Future Circular Collider would cost £12 billion; with…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to this update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

One week until the Christmas break: where did 2023 go?

This week, DeepMind serve up proof that a large language model can create new knowledge.

Also, more news from the accelerating story that is the march of the humanoid robots. It’s clear next year will be a pivotal one for this technology.

And researchers hook up brain organoids to microchips to create a new kind of speech recognition system.

Let’s get into it!

🧮 Fun times at DeepMind

This week, yet another step forward in the epic journey we’ve taken with AI in 2023.

Researchers at Google DeepMind used a large language model (LLM) to create authentically new mathematical knowledge. Their new FunSearch system — so called because it searches through mathematical functions — wrote code that solved a famous geometrical puzzle called the cap set problem.

The researchers used an LLM called Codey, based on Google’s PaLM 2, which can generate code intended to solve a given maths problem. They tied Codey to an algorithm that evaluates its proposed solutions, and feeds the best ones back to iterate upon.

They established the cap set problem using the Python coding language, leaving blank spaces for the code that would express a solution. After a couple of million tries — and a few days — the mission was complete. FunSearch produced code that solved this geometrical problem, which mathematicians have been puzzling over since the early 1970s.

DeepMind say it’s the first time an AI has produced verifiable and authentically new information to solve a longstanding scientific problem.

‘To be honest with you,’ said Alhussein Fawzi, one of the DeepMind researchers behind the project, ‘we have hypotheses, but we don’t know exactly why this works.’

⚡ NWSH Take: For pure mathematicians, a solution to the cap set problem is a big deal. For the rest of us, not so much. But this result really matters, because it resolves a central and much-discussed question about LLMs: can they create new knowledge? // Until this week, many believed LLMs would never do this — they they’d only ever be able to synthesise and remix knowledge that already existed in their training data. But there was no solution to this problem in the data used to train Codey; instead, it created novel and true information all of its own making. This points a future in which LLMs solve problems in, for example, statistics and engineering, or can create new and viable scientific theories. // In other words, this little and somewhat nerdish research paper heralds a revolution. So far, only we humans have been able to push back the frontiers of what we know. It’s now clear that in 2024, we’ll have a partner in that enterprise. // For this reason and so many others, I’m increasingly convinced that an unprecedented socio-technological acceleration is coming. It’s been a wild year; things are about to get even wilder.

🤖 Like a human

A quick glimpse of two stories this week. Both point in one direction: the humanoids are coming.

Tesla released a new video of its humanoid robot, Optimus. The Generation 2 Optimus can do some pretty fancy stuff, including delicately handling an egg:

Meanwhile, researchers at the University of Tokyo hooked a robot up to GPT-4.

The Alter3 robot is able to understand spoken instructions and adopt a range of poses without those poses being pre-programmed into its database.

In other words, Alter3 is responding in real-time to natural spoken language; it’s an embodied version of GPT-4, best understood as a kind of text-to-motion model.

⚡ NWSH Take: The closing months of 2023 have brought a welter of humanoid robot news. Amazon are now trialling the Digit humanoid in some US fulfilment centres. The makers of Digit, Agility Robotics, are about to open the world’s first humanoid mass-production factory in Oregon. And the CCP says it plans to transform China’s economy via an army of these devices. Next year, then, will prove a pivotal one for the longstanding dream that is an automatic human. And Elon Musk wants Optimus to be the One Bot That Rules Them All. // The tricks we see Optimus performing in this new video are pre-programmed. But Tesla is building the world’s most capable machine vision AI via an unbeatable data set — funnelled to them from hundreds of thousands of on-road cars — and the world’s most powerful supercomputer for machine vision, Dojo. Agility Robotics stole an early lead by getting Digit inside Amazon warehouses. But longterm, it’s hard to see how anyone beats Optimus. // If humanoids are indeed imminent, some some big questions are looming. When humanoids outnumber people, says Musk, ‘it’s not even clear what the economy means at that point’. Next year, we’ll have to confront this prospect anew.

👾 Interface this

Also this week, some fascinating news on organoids and the future of human-machine interface.

Researchers at Indiana University Bloomington grew brain organoids — essentially clumps of brain cells — in a lab, and attached them to computer chips. When they connected this brain-chip composite to an AI system, they found it was able to perform computational tasks, and even do simple speech recognition.

Clips of spoken language were turned into electrical signals and fed to the brain-chip hybrid, which the researchers call Brainoware. The researchers found that the Brainoware was able to process these signals in a structured way and feed back signals of its own to the AI system, which decoded them as speech.

Lead scientist on the project, Feng Guo, says the result points to the possibility of new kinds of super-efficient bio-computers.

⚡ NWSH Take: Welcome to the weird — and somewhat terrifying — world of organoids. It’s only a week since I last wrote about them; they’ve become a NWSH obsession. I can’t understand why they’re not getting more attention; last year brain organoids taught themselves to play the video game Pong, ffs. // Okay, I’ve calmed down. We’re a long way from viable technologies here. Culturing brain organoids, and then sustaining them long enough and in large enough numbers to do anything useful, is extremely hard. But in the Pong story and this week’s Brainoware news we see a new form of human-machine interface blinking into fragile life. We see, too, a future in which we’re able to grow more computational power in the lab. This story is sure to evolve; I’ll keep watching.

🗓️ Also this week

🧠 Researchers at Western Sydney University say they’ll switch on the world’s first human brain-scale supercomputer in 2024. The DeepSouth computer will be capable of 228 trillion synaptic operations per second, around the same as that believed to take place in the human brain. The researchers say DeepSouth will help us understand more about both the brain, and possible routes to AGI.

⚖️ UK judges are now allowed to use ChatGPT to help them craft their legal rulings. New guidance from the Judicial Office for England and Wales says ChatGPT can be used to help judges summarise large volumes of information. The guidance also warns about ChatGPT’s tendency to hallucinate.

🌊 New research shows that frozen methane under ocean beds is more vulnerable to thawing than previously believed. Methane is a potent greenhouse gas; the researchers say the methane frozen under our oceans contains as much carbon as all of the remaining oil and gas on Earth. If released, this methane could significantly accelerate global heating.

🚗 Tesla has recalled more than 2 million cars after the US regulator found its Autopilot system is defective. The recall applies to every car sold since the launch of Autopilot in 2015. But this is a ‘recall’ in name only; Elon Musk says Tesla will push a software update to fix the issue, so that no cars need to be returned to Tesla.

🖼 The new WALT video generation model can crea…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to this update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

It’s a bumper instalment this week; what do we have in store?

Google DeepMind owned this week’s tech headlines with the release of Gemini, a new multi-modal AI intended to outdo GPT-4.

Meanwhile, Harvard researchers have created tiny biological robots that can heal human tissue.

And the world’s largest nuclear fusion reactor is now online in Japan.

Let’s go!

Gemini has liftoff

This week, major news out of Google’s DeepMind AI division.

The DeepMind team announced Gemini, a multi-modal LLM that looks to have pushed back the frontiers when it comes to these kinds of AI models.

Launch videos suggest Gemini can speak in real-time (though as I go to press doubts about that are being raised; more below). It understands text and image inputs, and can combine them in novel ways. Here it is giving ideas for toys to make out of blue and pink wool:

It can write code to a competition standard. In tests it outperformed 85% of the human competitors it was compared against; that means it’s excellent even when compared to some of the best coders on the planet.

Gemini can even perform sophisticated verbal and spatial reasoning, and handle complex mathematics. Imagine if you’d had this to help with your homework:

This is significant; OpenAI’s GPT-4 is notoriously bad at maths and logic puzzles.

And Google are, of course, taking direct aim at OpenAI with this launch. Gemini comes in three variants: Ultra, Pro, and Nano. US users can access the Pro version now via Bard, and the Ultra model will soon be made available to enterprise clients.

⚡ NWSH Take: It will take time to independently verify the claims DeepMind are making; there are some murmurs that their launch videos overstate Gemini’s competence. Still, there’s no denying this model looks impressive. // Scratch the surface, meanwhile, and we can discern some underlying signals about the future development of LLMs. This AI outperforms GPT-3.5 when it comes to linguistic tasks such as copy drafting. But it’s the multi-modal nature of Gemini that’s really significant; in particular, its ability to reason. LLMs are trained to do next word prediction; that means they’re brilliant at sounding right. But they lack any underlying ability to know whether what they’re saying is right, or even makes sense. Gemini seems to address this shortcoming. The promise of an LLM that can act as a true reasoning partner is exciting, should haunt the dreams of all at OpenAI. // OpenAI’s reported work on the still-mysterious Q* algorithm is also believed to be about reasoning. All this suggests we’re hitting the limits of the performance improvements to be gained simply by training LLMs on even larger data sets. Instead, the future belongs to those who can weave multiple models together. // Finally, a word for Alphabet’s CEO Sundar Pichai: kudos. Alphabet AI engineers invented the transformer model; then the company went missing. Gemini puts Alphabet firmly back in the race. And given the recent fiasco at OpenAI, Pichai this week looks like a man playing a canny long game. It’s going to be a fascinating 2024.

🤖 Anthrobots are go

Two stories this week signal powerful new avenues of discovery for the life sciences.

Scientists at Harvard and Tuft’s University have created tiny biological robots, called anthrobots, made out of human cells. In tests, the anthrobots were left in a small dish along with some damaged neural tissue. Scientists watched as the bots clumped together to form a superbot, which then repaired the damaged neurons.

Each anthrobot is made by taking a single cell from the human trachea. Those cells are covered in tiny hairs called cilia. The cell is then grown in a lab, and becomes a multi-cell entity called an organoid. In this case, the scientists created growth conditions that encouraged the cilia on these organoids to grow outwards; they then become something akin to little oars that allow the entity to move autonomously. And lo, an anthrobot has been created.

The researchers say that in future anthrobots made from a patient’s own cells could be used to perform repairs or deliver medicines to target locations.

Meanwhile, researchers at New York University created biological nanobots capable of self-replication. The bots are made from four strands of DNA, and when held in a solution made of this DNA raw material they’re able to assemble new copies of themselves.

⚡ NWSH Take: Organoids have long been a NWSH obsession. This work on anthrobots builds on the research — by the same team — that created xenobots, which I wrote about back in December 2021. And who can forget the brain organoids that taught themselves to play Pong, which I covered in October of last year? // The original xenobot researchers at Harvard and Tufts were startled when their bots first began to work together in groups, self-heal, and self-replicate. But xenobots are made out of frog cells, and so have limited applications when it comes to humans. Anthrobots, on the other hand, are human in origin. Given their ability to heal other tissues, they show immense promise when it comes to new medical and wellness treatments. // As so often at the moment, machine intelligence underpins these advances. To create the original xenobots, AI supercomputers were used to ‘simulate a billion year’s worth of evolution in just a few days’. No wonder Nvidia CEO Jensen Huang says ‘digital biology’ will be a central part of the AI story over the coming years. I’ll keep watching.

💥 Come together right now

The world’s largest nuclear fusion reactor came online in Japan this week.

JT-60SA, in the Ibaraki Prefecture, is an experimental reactor capable of heating plasma to 200 million degrees celsius. Scientists say it offers the best chance yet to test nuclear fusion as a source of near-infinite clean energy.

In fusion, two or more atomic nuclei are smashed together such that they become one; this results in an energy release.

Meanwhile, UK-based Rolls Royce showcased a prototype lunar nuclear fusion reactor, which they say could power a permanent human settlement on the Moon.

⚡ NWSH Take: Fusion is the energy dream that has remained, so far, just out of reach. It doesn’t output CO2. It doesn’t create a lot of dangerous nuclear waste, as fission does. And proponents say it could mean near-infinite renewable energy, on tap. // And now, we’re getting closer. Last year saw the first controlled fusion reaction that generated more energy than was needed to make the reaction happen: this is the longstanding net energy gain goal. And now a startup ecosystem is flourishing; US-based Helion, for example, are working to build the world’s first commercial fusion reactor. And they’ve laid down a clear timeline: the startup recently signed a deal with Microsoft to supply the tech giant with energy starting in 2028. // It remains to be seen whether Helion, or anyone else, can achieve fusion in this decade. But if someone does, it will be a transformative moment; and we’re closer than ever.

🗓️ Also this week

🧮 IBM announced Quantum System Two, its most powerful quantum computer. The system integrates three 133-qubit Heron processors. IBM also announced Condor, a new 1,000-qubit processor. IBM are leading the way, right now, towards useful and utility-scale quantum supercomputers. If that promise is realised it will unlock insane new capabilities across climate simulation, the creation of new medicines, supply chain management and more. Read an interview with IBM’s director of quantum, Jerry Chow, here.

🖼 Stability AI’s new image generator can create 150 images per second. StreamDiffusion is built on top of Stability AI’s sd-turbo image generation model. And X users are using it to create tens of thousands of cat pictures.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to this update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

This week, more AI magic rains from the sky.

Also, average temperatures on planet Earth exceed the 2C warming threshold for the first time.

And my take on the OpenAI fiasco. In the end, it’s about power.

Let’s get into it.

✨ Like magic

This week, further glimpses of the ongoing collision between human creativity and machine intelligence.

Stable Diffusion released Stable Video Diffusion, a new text-to-video model that looks to be a step beyond anything we’ve seen so far.

In keeping with the company’s open source mission, the code for the model is available at its GitHub repository.

Meanwhile, X users went wild for a new tool, Screenshot to Code, that leverages GPT-4 and DALLE 2 to take a screenshot of any web page and automatically write the code that will render it:

And Elon Musk announced that X’s new on-platform large language model, Grok, will launch to all Premium users next week:

Grok is trained on a vast dataset of X posts; it’s sure to be expert in writing posts with a great chance of going viral. What’s more, it will have access to X posts in real-time; that could make for a whole new way to discover and interact with news stories.

⚡ NWSH Take: This gallery of the week’s AI wonders could go on far longer. I didn’t mention the new voice-to-voice model from UK-based Eleven Labs, for example: just upload your own voice and hear it converted to that of a famous celebrity, or a custom character that you create. // What’s the broader point here? A couple of weeks ago I shared an excerpt from a long AI essay called Electricity and Magic. That essay argues for a two-sided model of machine intelligence and its manifestations in the coming decades. First, machine intelligence is becoming something foundational — akin to a form of fuel that will power an army of autonomous vehicles, robots, and more. But in our daily life AI will manifest differently; not as fuel, but as magic. The innovations above give a glimpse of what I’m talking about. AI is moving into domains — from music, to film-making, to writing — once believed to be impervious to encroachment by automation. It’s as though someone has waved a magic wand over our machines. // The crucial point to understand, though, when it comes to AI magic? The result won’t be, as many people imagine, the devaluation of human creativity. Instead, amid a tsunami of machine-generated outputs, what is uniquely human — including creative work grounded in embodied experience — will only become more prized.

🌊 Crossing over

Another significant, and unwelcome, climate milestone was passed in the last seven days.

According to the EU’s Copernicus Climate Change Services (CS3), Friday 17 November was the first day on which average global temperatures were more than 2C above pre-industrial levels.

Data for 17 November indicated that global surface air temperatures were 2.07C above those in 1850. Provisional data for the following day indicated a 2.06C elevation.

This doesn’t mean that the much-discussed 2C threshold has been crossed. For that, we’d need to see a sustained elevation above 2C.

CS3 is part of the EU’s Copernicus Earth Observation Programme, which draws on vast amounts of satellite and other data to track the changing planetary environment.

⚡ NWSH Take: It’s expected that we’ll see occasional 2C+ days well before we exceed the 2C limit as commonly defined. Still, this week saw both the first ever and the second day that global average temperatures tipped over the threshold. It’s pretty clear where we’re heading. // This news comes on the eve of the UN COP28 summit in Dubai, which starts on 30 November. Many view last year’s summit, held in Egypt, as the moment at which the internationally agreed 1.5C target slipped out of reach; the summit notably failed to agree on a phase out of all fossil fuels, despite support for that proposal from over 80 countries. But the summit did achieve something: the establishment of a Loss and Damage Fund intended to transfer tens of billions to developing nations most at risk from climate change to help them mitigate the impacts of floods, droughts, and more. // At COP28, expect another push for a commitment to phase out all fossil fuels. And expect petrostates — including the host — to resist that call. As consensus grows that the 2C target will be breached, more attention will turn to plans for adaptation — and who should pay for them.

Form an orderly Q*

I can’t let this instalment pass without talking about the OpenAI fiasco.

Tech watchers everywhere munched their popcorn this week while OpenAI proceeded to fire CEO Sam Altman and hire a new CEO, only to get rid of that new hire and rehire Altman five days later.

It’s still unclear what led the OpenAI board to eject Altman in such dramatic style. But the mainline theory is that this was about internal division between those who want prioritise the original and nonprofit mission to research safe machine intelligence, and those — Altman apparently among them — who want to move fast and make lots of money.

Yesterday, news agency Reuters made waves with claims that the debacle may have been related to an advance called Q*. The details of that advance — or indeed if there has been any advance — are unconfirmed. Cue a whole new wave of speculation:

As per the above, most believe Q* is related to a generalised form of q-learning — a kind of reinforcement learning — that would enable LLMs to solve multi-step logic problems. Or, in simpler terms, to take multiple and reasoned steps towards a long-range goal in the way we humans do all the time.

Reuters imply that this advance prompted some in the organisation to fear that OpenAI was getting (dangerously) close to Artificial General Intelligence. And that this is what sparked all the drama.

⚡ NWSH Take: It’s believed that OpenAI will start to train GPT-5 next year. If that is true, and if Q* really is a big step towards generalised agents, then the AI story will only accelerate across next 12 months. We’re all, by now, accustomed to tech hype cycles (the metaverse!) but it’s becoming ever-harder to deny that something significant is happening. // But the events of this week also make clear another truth. Some technologists, including Altman, want us to believe that this technology is so powerful that we may lose control of it entirely, with existentially bad results for humanity. My hunch is that this is something of a psyop, designed to distract us from the real danger: AI that is controlled, but by a tiny, unaccountable, and chaotic group of Silicon Valley technologists. // At the heart of this is an an eternal aspect of human affairs that techno-accelerationists rarely want to discuss: power relations. Who gets to control this transformative new force, trained on a literary and cultural legacy that belongs to us all? Sam Altman? The OpenAI board? It seems the move fast and make money contingent at OpenAI won this battle; but should that be the end of it? Altman has waged a long marketing campaign around the idea that the AI he’s developing is powerful enough to pose existential risks. This feels a good time to call his bluff on that. Will he tell us what happened inside OpenAI across the last seven days? If not, perhaps we should send in public representatives to discover the truth.

🗓️ Also this week

👨‍💻 A former Googler made headlines with a resignation note that claimed morale inside the company is at ‘an all-time low’. Ian Hickson worked at Google for 18 years; he says the organisation’s culture is ‘eroded’ and accused CEO Sundar Pichai of a lack of vision. Google AI engineers developed the transformer model that underpins the generative AI revolution, but the company has seen its AI efforts outshone by OpenAI and its partner Microsoft.

☀️ Portugal ran entire…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to the mid-week update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

This week, Microsoft and Nvidia go head to head with new chips intended to train the next generation of AI models. And a clever hoax underlines a powerful truth when it comes to the war for compute power.

Meanwhile, a viral tweet about viral TikToks engenders another viral tweet. The lesson here? We’re living in a deeply enweirdened informational environment.

And in a world first, the UK approves a CRISPR-fuelled medicine.

Let’s go!

👾 Compute wars

This week, a glimpse of an emerging power struggle set to help shape the decades ahead. This isn’t a battle for land or natural resources. I’m talking about the struggle for compute power.

Microsoft announced their first and long-awaited custom AI chips, the Azure Maia AI chip and Cobalt CPU. Set to arrive in 2024, the chips will power Microsoft’s Azure data centres, and are intended to train the next generation of large language models (LLMs).

And Nvidia launched its new H200 AI chip, the successor to the H100. The iconic H100 is the fuel that’s driven this AI moment; huge clusters, consisting of tens of thousands of H100s, were used to train pretty much every large AI model you can name, including GPT-4.

Meanwhile, something quite different. A mysterious company called Del Complex announced the BlueSea Frontier Compute Cluster: a massive offshore data centre intended to circumvent the new the US Executive Order that says organisations training the most powerful new AI models must share information with government.

Del Complex calls BlueSea Frontier ‘a new sovereign nation state’. The announcement post achieved 2.5 million views, and was accompanied by a fancy website featuring images of BlueSea scientists at work. Tech blogs reported on the launch.

But wait; it is all a hoax! BlueSea Frontier is a comment on the These Strange Times by an artist and developer called (or so he claims) Sterling Crispin.

But I think Crispin may be onto something.

⚡ NWSH Take: The Del Complex hoax was a great bit of online trickery. But it was so convincing because it taps into a deep underlying truth. Compute is becoming a crucial nexus for techno-economic, sovereign, and geopolitical power. // The tech battle taking shape here is just one dimension of a broader story. Microsoft need to supply huge compute resources to their partner OpenAI to allow it to fully commercialise ChatGPT and train the upcoming GPT-5. So far, their data centres have been dependent on Nvidia AI chips. The new Maia AI and Cobalt CPU chips are intended to change that. // The broader story? It’s now clear that those nation states with the best machine intelligence will own the geopolitical future. The USA and China are now locked in a race to develop the vast compute needed to develop ultra-powerful next-generation models. Last year’s US CHIPS Act devotes $280 billion to semiconductor and AI research; inflation adjusted that’s more than the cost of the entire Apollo moon programme. And last week I wrote about new US restrictions on chip exports, intended to hamper China’s AI efforts. // It wouldn’t surprise me, then, if we do see the establishment of new offshore compute clusters, or even the development of new pseudo-sovereign entities based around compute power and AI. As with all the best satire, Del Complex’s vision is so wild it might just come true.

🔍 Can’t handle the truth

Also this week, another reminder of the hall of mirrors that is our new and connected media environment.

US journalist and X (formerly Twitter) personality Yashar Ali went viral with a tweet about TikTok. Ali claimed that across the previous 24 hours, many thousands of TikToks had been posted in which mostly young north Americans claimed to have read and agreed with Osama bin Laden’s notorious 2002 ‘Letter to America’ manifesto.

In the comments, theories abounded. Some said it was a signal of gen Z’s misguided politics. Others saw conspiracy, and said it was another indication that China is using TikTok as a channel for sophisticated psyops intended to destabilise the Global North. We should, said those people, ban TikTok.

Then another X user went viral with a different idea. These Bin Laden TikToks were being made and seen in huge numbers, he said, only because of Yashar Ali’s original tweet.

Other people said that was stupid, and itself tantamount to a conspiracy theory.

Meanwhile, this week the EU decided it would stop all advertising on X due to ‘widespread concerns relating to the spread of disinformation’. This follows EU research published in September which concluded that X is now the biggest online source of disinformation.

⚡ NWSH Take: Is TikTok an app for fun dance memes or a highly sophisticated channel for Chinese cultural warfare? Is the X algorithm now giving higher priority to toxic content, or is that just anti-Elon paranoia? Did thousands of young north Americans organically discover and agree with the Bin Laden letter, or is a dark controlling force at work? // The answer in every case: no one knows for sure. And that in itself is an indication of where we’re at. // The information environment that mediates our democracies has become insanely fragmented and opaque. The world’s richest man has total control over a key global information channel. The CCP has its hands around another. In both cases, I find it impossible to believe that the parties in question aren’t up to some tricks. // A totally connected world, in which every individual is empowered with a voice of their own, was supposed to create information nirvana. Those who bought that idea couldn’t have been more wrong. We need old media principles — editorial standards and, yes, gatekeepers — more than ever. But millions in the global north are currently convinced that the New York Times and the BBC are the real problem. In this increasingly chaotic and paranoid information environment, those institutions and others like them need to adapt rapidly. Most of all, they must rejuvenate belief in what they offer.

🧬 Major edits

Huge CRISPR news this week.

The UK’s medicines regulator became the first in the world to approve a medical treatment that uses CRISPR gene editing technology. The medicine, Casgevy, is a treatment for sickle cell disease, a serious inherited disorder that causes red blood cells to malfunction and that affects millions worldwide.

During treatment, red blood-producing stem cells must be taken from the patient. CRISPR is used to edit those stem cells to remove the error that causes sickle cell, before the edited cells are infused back into the patient.

Meanwhile, researchers at the Chinese Academy of Sciences created a monkey using two embryos, with donor material from one embryo injected into another. This has been done before with simpler animals such as mice and rats, but is a first in primates.

The donor stem cells were gene edited to express a green fluorescent protein, causing the resultant live monkey to glow:

⚡ NWSH Take: Gene editing technology is already enacting a transformation in the life sciences, healthcare, and agriculture. This CRISPR sickle cell treatment is wonderful news, and there are promising early indications from trials of CRISPR therapies to cure a form of hereditary blindness, and to train immune cells to fight certain cancers. Meanwhile, in September 2021 Japanese startup Sanatech Seed became the first company to sell CRISPR-edited food: their tomatoes were edited to contain more GABA. // So we’re developing our ability to manipulate genes. The next revolution coming? That ability will collide with a new ability to speak the language of DNA via transformer models — the kind of models that underlie LLMs — trained on…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to the mid-week update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

It’s a bumper instalment this week. What do we have in store?

The Chinese government is calling on its technology industry to roll out millions of advanced humanoid robots.

Also, NASA wants to learn how to extract breathable oxygen from Moon dust. And OpenAI says everyone can now create their own bespoke version of ChatGPT.

Let’s go!

🤖 Work machines

This week, a glimpse of the coming collision between human population dynamics and autonomous machines.

A new study by researchers at University College London found fears of climate breakdown are changing decision-making around whether or not to have children. Published in the journal PLOS Climate, the research found that climate concern was associated with a desire for fewer children, or none at all.

The researchers say theirs is the first systematic study of the way attitudes to climate change are affecting reproductive choices.

Meanwhile, the Chinese Ministry of Industry and Information Technology (MIIT) issued a nine-page communique calling for domestic mass production of advanced humanoid robots by 2025. By 2027, the document says, these robot workers should be ‘an important new engine of economic growth’.

But what is the connection between new trends in reproductive decision making and China’s dash towards humanoid robots?

Here’s a graph of the birth rate in China from 2000 to 2022:

⚡ NWSH Take: The CCP knows that China is losing its battle with demographics. If the country is to become the 21st-century hegemon that President Xi dreams about, then it needs an army of workers. But instead China is watching its birth rate plummet. Meanwhile, the Global North is facing the same challenge; in north America and western Europe population growth flatlined long ago. And now it seems that fears over climate change are only set to exacerbate that trend. // This is a huge structural challenge; fewer workers tends to mean a less productive and smaller economy. So what to do? The CCP have already tried ditching the one child policy and incentivising couples to have more children; it didn’t work. This week’s clarion call from the MIIT offers us a glimpse of an alternative answer: robots. If China won’t have enough human workers to sustain economic growth, then the CCP hopes humanoid robot workers can do the job(s) for them. // Innovators in the Global North are heading in the same direction. This week, Tesla posted over 50 jobs ads for its Optimus robot team. Elon Musk — who has long bemoaned population decline and its coming impacts — has said he believes Optimus will end up being a bigger part of Tesla’s business than EVs. And two weeks ago I wrote on how Amazon are trialling the Digit humanoid robot in some US fulfilment centres. // My co-founder at The Exponentialist, Raoul Pal, says that in the new world we’re building robots are demographics. In other words, the rise of autonomous machines is set to decouple economic growth from population growth. The CCP, Musk, and many others besides are making the same bet. And my guess? They’re going to be proven right.

🌌 Space out

NASA continues to prepare for its mission to the Moon. This week, further news.

The Agency wants to explore methods to extract breathable oxygen from Moon dust. Its Space Technology Mission Directorate is seeking input from industry partners and external researchers, and hopes to create a demonstration technology soon.

NASA hopes to put humans back on the Moon for the first time since 1972 with its Artemis 3 mission, currently planned for 2025.

Meanwhile, stunning pictures came back this week from the European Space Agency’s Euclid telescope. Launched in July, Euclid is now around 1.5 million kilometres from Earth; that’s about four times as far away as the Moon. And it’s capturing images of incredible clarity.

This is the Perseus cluster, a group of over 1,000 galaxies located 240 million light years from Earth. Each galaxy pictured — and there are a further 100,000 galaxies in the background of the shot — contains hundreds of billions of stars:

Here’s the Horsehead Nebula, a cloud of dust and gas in the Orion constellation:

⚡ NWSH Take: Okay, this entire segment was mainly an excuse to show you the breathtaking images coming back from Euclid. But there is an underlying truth here. We’re amid a new space age, due mainly to the insane drop in the cost of access to space. Back in 2010 launch costs hovered at around $20,000/kg; today they’re around $1,000/kg. That’s thanks mainly to the reusable rocket technology developed by SpaceX. We’re heading back into space via multiple partnerships between the international space agencies and private companies. And this time the plan is to stay there. // One signal of the emerging public-private space ecosystem? This week, SpaceX agreed to deliver the US military’s new space plane, the X-37B, into orbit in its Heavy Falcon rocket in December. And private space companies, including SpaceX, will play a huge role in the upcoming Artemis crewed mission to the Moon. Most analysts reckon that mission will end up being delayed until 2026/7. Even so, the next few years are set to be a thrilling road towards the lunar surface. Expect Moon hype to reach fever pitch. And from there, of course, all roads will lead to Mars.

🧠 Your intelligence

There’s little doubt about the biggest story in the mainstream tech press this week.

OpenAI made headlines all over again with the launch of custom GPTs: bespoke versions of ChatGPT that any user can create using simple natural language instructions and their own training content or data.

The feature was announced at OpenAI Dev Day, which saw CEO Sam Altman create a custom Startup Mentor GPT live on stage in about five minutes.

X (formerly Twitter) went wild. And yes, a million and one GPTs are assuredly coming.

How is this going to play out?

⚡ NWSH Take: Remember back in 2012, when every third friend of yours was making an app? OpenAI are hoping to recreate that magic all over again. They want to be the platform that profits from a huge wave of AI innovation. ChatGPT Plus users will be able to create custom GPTs and charge others for use, and Altman say they’ll be rewarded via revenue share. // Remember, any ChatGPT Plus user can now create a bespoke GPT in a few minutes. There will be a vast long tail of these things. The winners, though, will be those with (i) deep reserves of proprietary content or data that they can use to enhance the outputs of their bot, and (ii) audiences who are receptive to their creations. // But creating a bespoke GPT is now so easy that we’ll also see something we didn’t with apps. That is, individuals creating bespoke bots just for their own use — to help them manage their accounts, or choose birthday presents for family and friends, and much else besides. Yes, this is an App Store moment for AI. But it also marks another beginning: of personalised machine intelligence on tap.

🗓️ Also this week

💥 The Exponentialist, my new premium and enterprise-level research service, launched to the world! It’s a partnership between me and the macro economist and Real Vision CEO Raoul Pal. To mark launch day, we’ve made an excerpt of the first essay free for all to read — watch out for it in your inbox on Sunday.

📌 New tech company Humane launched the AI Pin. This long-awaited first product from Humane is a voice and gesture-controlled device that clips to your shirt and integrates with ChatGPT and other services. Humane hope their ‘disappearing computer’ will be the next iPhone. It remains to be seen whether people really want to talk to a badge on their lapel. One fascinating signal, though? See how OpenAI — and their partner, Microsoft — are set to become the underlying infrastructure that fuels a whole raft of AI innovatio…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to the mid-week update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

This week, two intriguing stories and a big announcement.

Global leaders and senior tech executives gathered at the UK’s AI Safety Summit. But beyond the walls of Bletchley Park, the debate on AI is raging hotter than ever.

Meanwhile, tech billionaires in Silicon Valley are running into trouble over their plans to build a new city-state utopia called California Forever.

As for the announcement? Just keep scrolling.

Let’s do this.

🧠 Dream machines

The UK government this week trumpeted the success of its international AI gathering; it took place at the historic fountainhead of the computer revolution, Bletchley Park.

An impressive guest list, including US vice-president Kamala Harris and the European Commission president Ursula von der Leyen, gathered at the Summit. And their meeting resulted in the Bletchley Declaration, which the UK government has hailed as a world-first international statement on AI safety.

Here’s a taste for those who speak technocrat:

‘We affirm that, whilst safety must be considered across the AI lifecycle, actors developing frontier AI capabilities, in particular those AI systems which are unusually powerful and potentially harmful, have a particularly strong responsibility for ensuring the safety of these AI systems…We encourage all relevant actors to provide context-appropriate transparency and accountability on their plans to measure, monitor and mitigate potentially harmful capabilities and the associated effects that may emerge…’

But beyond the Declaration, this week made it clear that we’re further than ever from a consensus on the deep implications of machine intelligence. In fact, this was the week that a maximum volume war of words broke out between leading AI builders.

Google Brain co-founder and Stanford professor Andrew Ng said key AI players, including Sam Altman, are wildly playing up fears of AI doom in order to spark regulation that will suppress competition from insurgents. He called the proposal that the training of powerful AI models should require a license ‘colossally dumb’.

That message was echoed by Meta’s chief AI scientist Yann LeCun, who favours open source — that is, anyone can use it — AI models.

But Google DeepMind CEO Demis Hassabis hit back at LeCunn, saying that failure to regulate AI could result in ‘grim’ consequences for humanity.

This account barely scratches the surface of the arguments that raged this week.

As for OpenAI, they launched a new team intended to study and prepare for ‘catastrophic risks’ including an AI-instigated nuclear war.

⚡ NWSH Take: Who would ever have thought that a bunch of super-smart, tech-obsessed social media addicts would end up arguing like this? While Bletchley saw a rare moment of diplomatic unity, inside the AI industry the full spectrum of opinion is manifest, from AI doom is all a load of rubbish to act now or the end of humanity is probable. // It pays, here, to remember that two things can be true at once. Yes, Altman’s global tour to warn of ‘catastrophic risks’ is a carefully orchestrated marketing campaign. But it’s also the case that no one, yet, has a definitive picture of the risks in play. // What is increasingly clear, though, is that the rise of machine intelligence is the primary fact of our shared lives now. It will do more than any other force to reshape our collective future. // But the Bletchley Declaration consists of bromides that will change nothing. And the sight at Bletchley this week of UK prime minister Rishi Sunak interviewing Elon Musk — positioning Musk as the star and Sunak as a fan — spoke volumes about the power imbalances we’ve allowed to evolve when it comes to government (i.e. the people) and unaccountable tech overlords. // We must recover our collective agency; our ability to assert human modes of living and being in the face of an ongoing technology revolution. That means doing politics. Bletchley was a start. But what’s needed next are citizen assemblies, and an authentic movement around AI for the people.

💥 The Exponentialist

As some of you will have seen on social media, I made a big announcement this week.

I’ve partnered with Raoul Pal, renowned macro-economic thinker and CEO of Real Vision, on a new premium research service called The Exponentialist.

This is a professional and enterprise-level service for those who want to go deep on emerging technologies, the futures they’ll create, and the challenges and opportunities latent in all that.

This won’t be for everyone in the NWSH community. But if you’re a foresight professional, strategist, founder, marketing leader, product manager, designer or much else besides, The Exponentialist will fuel you and your team. And it will take up only a fraction of your research budget.

It will also be deeply valuable for anyone seeking to position an investment portfolio around tech and crypto.

This launch changes nothing about New World Same Humans and the community we’re building here. Our mission continues unchanged!

If The Exponentialist sounds useful, go here to learn more. And if you’ve subscribed or you’re considering it, hit reply to this email so I can say thanks.

🏙 Now and Forever

While the newsletter was on pause, we learned that a group of Silicon Valley billionaires are planning a new city-state utopia in California. This week, it seems their project has run into trouble.

California Forever is a new city planned for construction in Solano County in the north of the state. It’s backed by some of tech’s most notable power players, including ultra-rich VC Marc Andreessen, Stripe founders Patrick and John Collison, and LinkedIn founder Reid Hoffman.

The group’s vision for the city has strong solar punk, hi-tech sustainable utopia vibes:

But this week it was reported that the mysterious company behind the plans, Flannery Associates, is accused of using ‘strong-arm tactics’ including lease terminations to buy up the Bay Area farmland it needs. Local farmers aren’t happy, and now some of them are taking the matter to court.

Trouble in (planned) paradise, then.

⚡ NWSH Take: This project reminds me of the various other pseudo-independent city-states discussed in this newsletter over the years. There’s Walmart billionaire Marc Lore’s Telosa City, for example, a sustainable paradise planned for the Nevada desert. And Praxis, a startup on a mission to build a new Great City somewhere in the Mediterranean, funded by NFTs of the monuments they’ll build in the city once it exists. // Few details have emerged of the way California Forever will be governed. But for a glimpse, we might turn to billionaire backer Marc Andreessen’s recent’s Techno-Optimist Manifesto, which proclaims: ‘we believe in ambition, aggression, persistence, relentlessness — strength.’ I’m thinking libertarian, with a strong emphasis on innovation and startup culture. // Of course, innovation and startups can be great. But they only function in the context of the broader socio-political frameworks that libertarians such as Andreessen repudiate. As with the other charter city projects covered in this newsletter, I can’t help feeling that at the heart of Forever California is a fantasy of permanent escape from politics. Escape, that is, from the messy, awkward business of managing conflict among different interest groups, and enacting trade-offs between different but equally legitimate value systems. This argument with the farmers might be the first public conflict that Forever California has run into, but it won’t be the last.

🗓️ Also this week

🎬 Hollywood actress Scarlett Johansson is suing an AI app for cloning her voice and using it in an advert. Johansson says Lisa AI: 90s Yearbook and Avatar used an AI version of her voice without permission. Last week I wrote on the coming wave of…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to the mid-week update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 25,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

It’s great to be back! Here, as promised, is the first post-break instalment of New Week. What do we have in store?

Nvidia are using an AI agent called Eureka to autonomously train simulated robots in virtual environments.

Meanwhile, research from Pew shows far fewer US adults are following the news; couple that with emerging deepfake technology, and 2024 should make for an interesting Presidential election year.

And a new AI model, 3D-GPT, can turn text prompts into amazing 3D worlds.

Let’s get into it.

🦾 Robot education

I’ve written often about Nvidia’s Omniverse platform: an AI-fuelled industrial metaverse that’s being used by BMW, for example, to simulate entire factories.

This week, Nvidia showcased Eureka, an autonomous AI agent that can be set loose on simulated robots and train them to perform complex tasks.

Eureka uses GPT-4 to write code that sets the simulated robots specific goals, and starts them on loops of trial and error learning. As the robot sets about its task Eureka will gather feedback and iterate its code, leading to a virtuous circle of better code and faster learning.

Via the agent, simulated robots inside Omniverse have learned to perform over 30 complex physical tasks. Including highly dextrous pen manipulation, handling of cubes, and opening doors:

Nvidia says that trial-and-error learning code generated by Eureka outperforms that created by human experts for over 80% of the tasks studied so far.

Meanwhile, Amazon this week trialled a humanoid robot called Digit in some of its US warehouses. The company says Digit could ‘free up’ warehouse staff to perform other tasks.

⚡ NWSH Take: There’s no doubt: the robots are coming. I laughed when Elon Musk announced the Tesla humanoid robot, Optimus, alongside a man dancing in a white spandex suit. Two years on, Optimus is autonomously sorting objects by hand. The pace of development has been insane. // Eureka and AI agents like it, though, have the potential to spark an explosion in robot competence. Teaching robots to navigate physical environments is hard. Now, we’ll be able to establish recursive loops of trial, error, and improvement in virtual space — no human input needed. // What could this competence explosion mean? When it comes to work, look to this week’s Amazon trial. Amazon employs 1.6 million people in its fulfilment centres worldwide, and currently it’s deploying the usual line: ‘these robots will free up staff, not replace them’. That’s hard to believe longterm; a phase of job displacement is coming, and it’s going to be painful for many. // Meanwhile, robots will make their way through workplaces and into our homes. Recently I spoke to legendary tech analyst Robert Scoble; he sees a future in which humanoid robots are delivered to homes on-demand by autonomous vehicles to vacuum, empty the dishwasher, and make the coffee. For further thoughts on that future, read Our Coming Robot Utopia.

📰 What news

This week, Pew Research gave a fascinating insight into our changing information environment.

A new survey shows that the proportion of US adults who closely follow the news has dropped precipitously across the last few years. Back in 2016, 51% of US adults said they followed the news all or most of the time. By 2022, that number had fallen to 38%.

Remarkably, the decline has taken place across all demographic lines, including age, gender, ethnicity, and voting preference.

⚡ NWSH Take: This feels like a big deal. We’re heading into a US presidential election year. And in 2024 a new set of circumstances are going to pertain. First, deepfakes are set to cause chaos as never before; just see this week’s convincing fake of Greta Thunberg in which she appears to call for ‘sustainable weaponry’. And now, via this research, we know that far fewer US voters are paying close attention to conventional sources of news. What happens to presidential campaigns in this kind of media environment? We’re going to find out. // Meanwhile, the longterm structural challenges are clear. Decades ago, the pioneers of Web 2.0 — I’m looking at you, Zuck — sold us on the idea that a connected world would mean a world informed and enlightened as never before. It hasn’t turned out that way. In fact, social media has turned many away from news as traditionally defined, and towards unverified gossip and conspiracy theory. The institutions and processes of our democracies evolved to function in symbiosis with an established media that operates under certain standards, and that is the primary source of information for voters. All that is now falling apart. Our democracies — what they are, how they work — are going to change. The 2024 presidential elections will be a window on to what is coming.

🗺 Hello world

This newsletter has watched the unfolding generative AI text-to-image revolution closely. But it’s always had one eye on another, even more compelling destination: text-to-worlds.

Now, that dream is being realised.

Researchers from the Australian National University, the University of Oxford, and the Beijing Academy of Artificial Intelligence this week showcased a new AI model called 3D-GPT. It generates 3D worlds based on natural language text prompts provided by the user.

According to the research paper, model deploys multiple AI agents to understand the text prompt and execute 3D modelling inside the open source modelling platform Blender.

See that paper for more on some of the worlds generated, including ‘A serene winter landscape, with snow covered evergreen trees and a frozen lake reflecting the pale sunlight.’

⚡ NWSH Take: 3D-GPT takes its place alongside this prototype text-to-world tool created by Blockade Labs, which I wrote about back in April. Where is all this heading? We’re still pretty deep in a metaverse winter right now, though there are signs of a thaw; the most obvious being the imminent arrival of the Apple Vision Pro mixed reality headset, which could act for millions as a gateway into more sophisticated virtual environments. While the word metaverse is probably damaged beyond repair, I still believe that immersive virtual worlds will play a role in our shared future. // What we’re talking about with text-to-world models, though, is even more head-spinning. 3D-GPT builds worlds that we look at via a screen. But eventually, we’ll be able to create entire, immersive, highly realistic VR worlds simply by describing them. In this way we’ll become something akin to sorcerers, able to confect new realities on command. That will transform video gaming and film. It will fuel new art forms and modes of collective expression. And, ultimately, it will change our relationship with reality — that is, with this reality — itself.

🗓️ Also this week

🎨 A new anti-AI tool allows artists to prevent AI models such as DALL-E from using their work as training data. Nightshade, dubbed a data poisoning tool, can be attached to creative work; if that work is scraped to be used to train an AI model, then Nightshade will corrupt the entire training database. We’re going to see a rising number of disputes between owners of creative IP and the owners of AI models who used that work as training material. See also, this week, Universal Music Group’s lawsuit against Anthropic; UMG say Anthropic unlawfully used its song lyrics to help train the Claude AI chatbot. And now major newspapers, including the New York Times, are seeking payment from OpenAI for use of their content to help train GPT-4.

☀️ The International Energy Agency says the global shift towards renewable energy is now ‘unstoppable’. The Agency’s latest World Energy Outlook report says renewables — mainly solar and wind — will provide half the world’s electricity by 2030.

🛰 NASA’s interstellar Voyager probes had a sof…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz

Welcome to the mid-week update from New World Same Humans, a newsletter on trends, technology, and society by David Mattin.

If you’re reading this and haven’t yet subscribed, join 24,000+ curious souls on a journey to build a better future 🚀🔮

To Begin

It’s a bumper instalment this week. What’s coming up?

Google showcase their AI arsenal at the annual I/O developers conference.

Meanwhile, new research reveals that the ocean currents may be about to take a weird turn, with disruptive results for the global climate. And a Snapchat influencer launches an AI version of herself to her 1.9 million followers.

Let’s go!

💥 The search empire strikes back

This week, Alphabet leaned hard into an AI everywhere, for everyone strategy at its I/O developers conference.

CEO Sundar Pichai announced PaLM 2, an update on the company’s primary large language model. Google’s Bard chatbot is now fuelled by the new model, and has been made available globally with no waitlist.

There was much more. Google execs announced new AI features in Maps, and a powerful new magic editor for photos that brings Photoshop-like capabilities into the phone. Pichai said AI around one zillion times, and Google later published a handy summary of all the announcements.

The centrepiece, though, was a demonstration of Google’s plans to weave generative AI through search. In this new search experience AI-generated results take up most of the first screen; users in the US can now access this experimental version of search via Google Labs.

The I/O conference wasn’t the only source of intriguing announcements from Google, though. The company also launched Geospatial Creator, an impressive tool that allows creators to build and publish geolocated AR installations. Essentially, to build a digital object and drop it anywhere on the surface of the Earth.

The tool is powered by the Google Maps platform, and integrated into Adobe Aero and Unity.

⚡ NWSH Take: Google researchers invented the transformer models that underpin this generative AI revolution. But across the last two years the tech giant has watched OpenAI steal its thunder. This week’s I/O conference was a statement of intent: we’re taking back control. Competition can only be good for users, many of whom will have gone straight to the new PaLM 2-powered Bard to compare it to ChatGPT. My anecdotal experience is that Bard is faster — ChatGPT with GPT-4 is a little slow — but the consensus at the moment is that it’s currently less factually reliable. Meanwhile Google is working on Gemini, a multimodal LLM clearly intended as a GPT-4 killer. The war for supremacy between Alphabet and OpenAI-Microsoft is just getting started. // Geospatial Creator was overshadowed by I/O, which feels fitting for a year in which the metaverse has been comprehensively out-hyped by AI. But the tool is intriguing glimpse of the emerging unified digital-physical field. Build a digital sculpture from your desk in London, and drop in into a park in São Paulo for your subscribers to view. And pretty soon, via text-to-everything models, you’ll be able simply to describe that sculpture and watch an AI model build it for you. A couple of years ago I wrote about the ways in which AR will change our relationship with a shared physical reality. I stand by those ideas, but in the age of generative AI that essay needs an update; one will be coming soon.

🌊 Climate weirding

Also this week, new research says changes in the ocean currents may soon enweirden the climate of northern Europe.

The Beaufort Gyre moves in a clockwise direction around the western Arctic Ocean, and helps regulate sea ice formation in that region. Scientists have long suspected that climate change is causing changes to the Gyre’s movement.

This new paper, Recent State Transition of the Arctic Ocean’s Beaufort Gyre, was published in Nature, and makes use of satellite data collected between 2011 and 2019. It provides the first observational confirmation that the Gyre is slowing and has entered a new ‘quasi-stable state’.

This means, say the scientists, that the Gyre may soon expel a massive amount of icy fresh water into the North Atlantic.

And that could spark further ocean current changes that cause the climate in western Europe to become significantly cooler.

⚡ NWSH Take: Yes, cooler. I’m no ocean currents expert, and I found this quick explainer on the Beaufort Gyre helpful. Essentially, the Gyre periodically sucks in a ton of icy fresh water and then exhales it, and it’s now long overdue an exhale; when that massive exhale comes it could send other ocean currents askew in ways that dramatically cool western Europe. Remember, the Gulf Stream — a major ocean current responsible for several global weather patterns — has slowed by around 16% already; scientists are scrambling to understand how a huge Beaufort Gyre exhale will impact this. // The upshot? One way or another, we’re probably about to undergo a climate weirding on a scale that few of us are ready for. While drought and fires rage in some places, a new freeze will break out in others. At the outer edges of this is the risk the Gulf Stream shuts down entirely, triggering rapid and chaotic climate disruption fuelled by a set of feedback loops. These processes are hugely complex; we’ll see much more work such as this attempt to build machine learning-fuelled simulations that give us advance warning of ocean current shift. Perhaps NVDIA’s coming and massive Earth-2 simulation can help.

❤️ Hey girlfriend

Regular readers know that virtual companions are a longstanding NWSH obsession. This week, another glimpse of what is coming.

Snapchat influencer Caryn Marjorie, who has 1.9 million followers on the platform, released an AI girlfriend version of herself. Users pay $1 per minute to chat to CarynAI, which the creator says is built on top of GPT-4 and trained on over 2,000 hours of her video and voice content.

Marjorie says the bot made $72,000 in the first week of release. She says that it could make around $5 million per month if 20,000 people — or just 1% of her Snapchat followers — subscribe.

So far, things seem to be going well:

⚡ NWSH Take: Back in 2013 I started telling leaders in big corporations that a new age of AI-fuelled conversational agents was coming. That people would even have 'relationships' with these new virtual entities; that it would be something way beyond Siri — their best reference point at the time. Some learned forwards; some raised a sceptical eyebrow. My constant refrain back then? I know it sounds like science-fiction, but it’s coming. Well, it’s here. Virtual Companions are set to unlock new manifestations of some of the deepest and most powerful human impulses: social connection, friendship, intimacy. // Observing this truth is not the same as celebrating it. What happens to authentic human connection in a world in which we simulate it — and commodify those simulations — in this way? What harms are we doing to vulnerable people who become attached to, even dependent on, these creations? // The central message still pertains: it’s weird but it’s happening. In the end I can’t help feeling that so much about contemporary living on the internet — the way it atomises our attention, the simulation of human relationships — must push us to finally realise that authentic human being together is the only sphere of activity invulnerable to technological advance. No machine can be a human, truly seeing you as another human. In the age of the machine, that truth becomes sacred.

🗓️ Also this week

⚛ Microsoft announced a partnership with fusion power provider Helion Energy. The deal will see Microsoft buy electricity created by a Helion fusion plant, which is expected to be operational by 2028; Helion says it marks the world’s first fusion power purchase agreement between two companies. Microsoft’s Azure Cloud platform will need vast amounts of compute power — using stupendous amounts of energy — given its commitment to support O…

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.newworldsamehumans.xyz