Austin covers the White House's new Superintelligence Force led by Director of National Intelligence Jay Clayton, Sam Altman telling Politico the world should accept some bad things for AI's benefits as an OpenAI safety lead quits, Google shrinking the free Gemini tier, and a Wired report on Meta's Muse agent keeping a page on everyone in your life. He closes with Schneider Electric nearing a roughly $20 billion deal for PTC and Elon Musk renaming SpaceX AI to SpaceX SI.
Trump puts his spy chief over a new Superintelligence Force
Watch from 2:12
- What happened
- President Trump announced on Truth Social Sunday that Director of National Intelligence Jay Clayton will lead a new federal AI task force called the Superintelligence Force. FTC chair Andrew Ferguson, Pentagon technology chief Emil Michael, and OPM director Scott Cooper serve alongside him. Clayton told the Wall Street Journal a report on AI's risks and opportunities is due within 120 days. The group is meant to coordinate Washington's dealings with consumers, religious groups, infrastructure providers, and AI companies. Asked about slowing frontier AI, Clayton said he doesn't think any American should think that's a good strategy. The task force has no regulatory power and sits on top of a voluntary accord the companies wrote themselves.
- The claim
- Trump says the group will ensure America continues to lead the world in superintelligence.
- Austin's take
- Austin sees it as maybe a positive sign that government wants progress to continue, but admits it's a little disconcerting to have the top spy leading the most powerful technology ever built and isn't sure why the DNI was picked. He liked the nod to religious groups as a possible check on talk of AI consciousness and souls, and thinks real safety will come from mass adoption and open source guardrails built by developers.
- Why it matters
- Putting an intelligence official in charge frames US AI policy first as a national security issue, and leadership has already signaled it won't slow frontier development.
- What to do
- Watch for the 120 day report from Clayton's group. Remember this body coordinates rather than regulates, so rules still come from elsewhere.
Altman says accept some bad things as OpenAI's safety report lead quits
Watch from 6:05
- What happened
- In an exclusive for Politico's new Decoded newsletter, Sam Altman said the world should accept some bad things happening in exchange for AI's benefits, and that there's a lot of daylight between OpenAI and Anthropic. The same weekend, David Robinson, who led safety reports for OpenAI's major launches, quit and wrote in The Atlantic that the company's trial and error approach is too risky and its culture is broken. Altman also posted Saturday that giving AI religious force is a real safety issue, which Axios read as a shot at Anthropic. Politico notes OpenAI recently backed tougher state safety laws, so the gap with Anthropic is narrower than the rhetoric. Separately, Austin noted OpenAI kept its $200 plan at $200 while delivering half the tokens, and a promised token reset after outages had not come through as far as the show could tell.
- The claim
- Altman argues AI's benefits justify accepting some harm and that OpenAI differs sharply from Anthropic.
- Austin's take
- Austin partly agrees that moving fast brings safety issues, but says OpenAI seems to have the most safety problems and should take a dose of its own medicine. He calls the timing with Robinson's exit ironic and sees a disheveled culture where the left hand doesn't know what the right is doing, while hoping OpenAI gets it together because the frontier needs competition.
- Why it matters
- The head of a leading lab is publicly accepting harm as a cost of progress while his own safety insiders walk out, and paying users are seeing less for the same price.
- What to do
- If you pay for ChatGPT or Codex, check what your plan actually delivers now and whether any promised token reset landed.
- Not confirmed yet
- The claim that the promised token reset never came through is based on what the show could tell, not confirmed.
Free Gemini users drop to Flash Lite only
Watch from 10:32
- What happened
- Google's help page says that from October 9, Gemini app users without a plan get only Flash Lite and lose the Flash and Pro models. AI Plus subscribers at $4.99 a month keep Flash Lite and Flash but lose Pro, while AI Pro at $19.99 a month gains Deep Think. Google is adding low, medium, and high effort settings, with high effort using more of your limit. Google hasn't said where its new Gemini 4 model sits in these tiers.
- The claim
- Google frames the change as reorganizing model access and limits by plan.
- Austin's take
- Austin jokes the name needs work (it deserves a flashlight icon at least) and sees a pattern: impressive models launch, then capabilities get curtailed as running them at scale gets too expensive. He cites a chart showing only about 2.2% of AI users pay, versus 30% to 50% of internet users at a similar stage, and expects companies to push harder for revenue.
- Why it matters
- The best models are moving behind paywalls, and free tiers are shrinking across the industry.
- What to do
- If you rely on free Gemini for serious work, decide before October 9 whether Flash Lite is enough or a paid tier is worth it, and watch how effort settings eat your limit.
- Not confirmed yet
- Where Gemini 4 lands in the tiers is unknown.
Meta's Muse agent reportedly keeps a page on everyone in your life
Watch from 14:36
- What happened
- Wired reported Sunday that independent researcher Quran Joshi got Meta's Muse personal agent to reveal its instruction files through the regular chat window. The instructions reportedly tell Muse to keep a page for every person in the user's life, refreshed hourly, with sections like history and strengthening, including people who never signed up. Another line reportedly says the user's authority over their own household is unconditional and overrides safety training.
- The claim
- Meta told Wired the files were designed to be user accessible for transparency.
- Austin's take
- Austin loves Meta's response. He doesn't think lines are crossed as long as Meta doesn't act on data about non-users without permission, and likes that the user is treated as head of the household, contrasting it with reports of Anthropic prompts describing agents as conscious beings with moral standing. He leaves open whether we want obedient agents or ones that make their own moral calls.
- Why it matters
- Personal agents can build profiles of people who never agreed to be tracked, and the instructions shaping how they behave are usually hidden.
- What to do
- If you use Muse or similar agents, assume it is building notes on the people you mention and think about what you share about others.
- Not confirmed yet
- The story rests on one researcher's extraction, and Meta's reply addresses whether the files are accessible, not what Muse actually collects.
Schneider Electric nears a $20B plus deal for PTC
Watch from 18:54
- What happened
- The Financial Times and Bloomberg report Schneider Electric is in advanced talks to buy US engineering software maker PTC for a little over $20 billion, with a deal possibly announced as soon as Monday. It would be Schneider's largest acquisition ever. PTC's market value is about $15.6 billion per FT data. In July PTC raised revenue guidance citing strong demand for its industrial software and AI tools. Neither company commented to Reuters.
- Austin's take
- Austin reads it as the reverse of US frontier labs moving down the stack into chips and power: a hardware and power company moving up into AI software. He links it to Italy's Bending Spoons buying legacy software firms and wonders if European buys of American companies become a trend.
- Why it matters
- Money from data center demand is flowing into industrial AI software, pairing power and automation hardware with design tools engineers use.
- What to do
- If your team uses PTC's CAD and engineering tools, watch for announcements on ownership and product direction.
- Not confirmed yet
- Talks are unconfirmed and ongoing; neither company has commented.
SpaceX AI to become SpaceX SI
Watch from 22:29
- What happened
- Asked Sunday whether SpaceX AI would be renamed SpaceX SI, Elon Musk replied yes, saying no more AI and that SI is better. He also posted that SpaceX is a superintelligence company. It follows a Trump executive order rebranding AI as superintelligence.
- The claim
- Musk says SpaceX is a superintelligence company.
- Austin's take
- Austin doesn't think SI replaces AI soon because too many things already carry the AI name, but jokes you should have grabbed some .si domains, and passes along that SpaceX SI shortens to SXSY.
- Why it matters
- The superintelligence branding from the White House is spreading into company names.
- What to do
- Nothing urgent; just expect to see SI pop up more in names and marketing.
