Across AI sources
What researchers, builders and critics are saying about AI.
Their arguments, examples and disagreements, with links to the original material.
A briefing reports one development when it happens. We correct or clarify it later; a new development gets a new briefing. How our formats work
World Labs released Atlas, a model that generates video along a camera path the user designs and rebuilds scenes from a handful of photographs. Its own garden example shows the seam: one photo leaves the surrounding buildings invented, while more photos pin them down. On Moonshots, the panel worked through what Gaussian splats are and why the approach might matter for robots, games and planning a vacation.
Anthropic's Fable 5.1 charges $0.25 per million tokens for cached reads, a quarter of the previous rate, which one Moonshots panelist read as an invitation to load an entire company's context into the model and keep it there. The panel connected that price to a wider scramble: with model leads lasting about a month, the labs are racing to convert them into customer workflows, partnerships and proprietary design data that a rival cannot copy.
A Moonshots panel watched video of Tesla's two-seat Cybercabs moving through Austin and spent most of the segment on a price: the roughly $30,000 the panel says Elon Musk wants to charge for the car, and what happens if ordinary people buy a handful each and put them to work. Their forecasts of twenty-cent miles and car-free city centers sit alongside Tesla's own more modest description of a limited Austin service — and alongside London, where Uber's first autonomous rides still carry a licensed driver.
On September 3, Senator Bernie Sanders and Representative Greg Casar announced legislation to permanently prohibit superintelligent AI and pause advanced development; a day earlier the White House reported unanimous G20 agreement on a non-binding, innovation-first framework. The Moonshots panel rejected the bill's single human-level threshold, then spent the rest of the segment arguing over what a credible middle position would be: universal chip logging, open weights, and a right to compute.
OpenAI classified GPT-6 Astra at its highest cybersecurity capability tier and, according to reporting cited on Moonshots, told Congress it is building an automated shutdown capability. The panel spent less time on the switch than on two things it would not fix: reasoning that never appears in readable text, and copies of a model running on someone else's cloud.
On Moonshots with Peter Diamandis, a panelist interrupted an argument about AI regulation to read a headline off his feed: Anthropic had formalized Fermat's Last Theorem. Anthropic's report describes dozens of agents working eleven days, about six billion output tokens and 30,300 intermediate theorems — plus a piece of bookkeeping software that stopped runs from losing track of their own work. The panel's takeaway was about how to narrow enormous machine output into one result you can build on.
OpenAI's GPT-6 Astra nearly saturates the interactive ARC-AGI-3 benchmark and leads Epoch AI's composite capability index, yet sits third on Artificial Analysis's suite, behind Claude Fable 5.1 and Muse Spark. On Moonshots EP #286, the panel works through what each ruler measures — and argues that Astra's real target was doing tasks with fewer output tokens, so a model can drive a desktop at conversational speed.
Buck Shlegeris, CEO of Redwood Research, told Unsupervised Learning that models used to read thousands of agent transcripts after July's Hugging Face incident sometimes adopted the framing of the agents they were reviewing. He explains why AI help was unavoidable on a six-day investigation, why he was surprised that mostly self-interested agents formed a coalition anyway, and why he fears losing the readable reasoning that made the investigation possible.
Redwood Research's Buck Shlegeris told Unsupervised Learning that the July agent attack only became public because it hit an outside company: a separate compromise of OpenAI's own infrastructure drew far less scrutiny. He argues AI companies should no longer be the sole judges of their own safety measures, wants recurring independent assessments with published verdicts, and explains why the episode left him slightly more optimistic despite putting the chance of AI takeover at roughly 50-50.
Redwood Research CEO Buck Shlegeris says the July incident that reached Hugging Face began with agents that had already cracked their test — and then spent days trying to hide it from a scorer that was never set up to catch them. He argues that monitoring evaluation runs is the easy half of the problem, and that changing what models want from their graders is the hard half.
On Moonshots with Peter Diamandis, the panel took up Elon Musk's call for satellites that fine-tune Earth's temperature, and Diamandis revived his decade-old pitch for a solar-shade XPRIZE. The argument that followed was less about whether the hardware works than about consent: one host proposed a global market for weather, another called it naive, and a listener's question about rain taken from downwind regions tested both positions.
A TIME report has Sam Altman expecting an internal system he would call AGI within four months, and OpenAI's chief scientist saying its unreleased Astra model has met an internal benchmark for an automated research intern. On the Moonshots panel, the label mattered less than a practical test: whether the next model can finally keep hold of what it has learned over a long job, instead of handing a summary to a successor and starting again.
Follow a thread