Austin tries a later-in-the-day format and works through six stories. Anthropic pledges 5% of the company to a public fund and Elon Musk answers with a SpaceX share pledge for every American kid. Also covered: Dallas garbage trucks with AI code-violation cameras, an eight-day chip stock slide, SoftBank eyeing a huge security guard company for robots, Claude explaining a 25-year-old NSA number mystery, and the NBA's new AI for spotting suspicious bets.
Watch from 2:28
- What happened
- Anthropic announced it is giving 5% of the company to a new fund for the public, estimated at about $48 billion based on its last valuation. Stockholders still have to approve the pledge. Anthropic says about 20% of the company now goes to the public good, and its own report says beginners are now 9% of its tech hires, down from 27%. Elon Musk replied on X recommending the shares go directly to the public, then pledged that next year he will give one SpaceX share to every kid in America, calling it the first of many. Musk gave no exact details.
- The claim
- Anthropic frames the fund as a way to pay people directly as AI changes how jobs are created and how people work.
- Austin's take
- Austin calls it a net positive and a great look for both Anthropic and Musk, but would rather see the money go to education and training that helps displaced workers use AI tools, not just subsidize losses. He also floats a more dystopian read, pointing to Dario Amodei's comments about Anthropic running everything, and wonders whether this ends with Anthropic deciding who gets money. He labels that speculation and hopes the companies follow through rather than just making headlines.
- Why it matters
- This is one of the first big, concrete attempts by an AI company to share its gains with the public as entry-level tech hiring shrinks, and it is already pulling other billionaires in.
- What to do
- Treat it as a promise, not money in hand. Watch for the stockholder vote and for details on who qualifies and how payouts work.
- Not confirmed yet
- The $48 billion figure is an estimate, stockholders have not approved the pledge, and Musk gave no specifics on how his share giveaway would work. A future value of up to $100 billion depends on an IPO valuation.
Watch from 7:16
- What happened
- Dallas garbage trucks carry AI cameras that flag homes for code violations. NBC 5 found the cameras flagged over 20,000 possible violations. City staff review the flags first, and from April to mid September the city sent about 3,700 notices and more than 80 citations. Each paid fine sends $36 to a fund that used to give low-income owners up to $20,000 for repairs, helping about 20 to 30 families, but those grants were paused in 2024. The city won't say how much money is in the fund.
- The claim
- The city says it may restart the fund.
- Austin's take
- Austin thinks the cameras are a smart, innovative use of AI in the physical world, but says the city has to connect the dots: you cannot use AI to collect faster and then not deliver on promises to residents. He expects early implementation hiccups and sees it as a net positive if the extra revenue actually helps people repair homes. He compares the pushback to Flock cameras.
- Why it matters
- AI lets a city inspect whole neighborhoods at once, so enforcement scales up fast while the help for owners who cannot afford fixes is stuck.
- What to do
- If you live in a city adopting AI enforcement, ask where the fine money goes and whether assistance programs are actually funded.
- Not confirmed yet
- It is unclear whether or when the fund will restart, and its balance has not been disclosed.
Watch from 11:20
- What happened
- Chip stocks fell eight days in a row, their longest losing streak since 2008. On Friday the chip index fell almost 4%, with all 30 of its stocks down. The morning of the show they bounced back almost 3% but were still down about 13% since October 6. Two worries drove the slide: heavy borrowing by AI companies and hotter-than-expected inflation. A big chip stock fund is still up about two thirds this year.
- The claim
- Nvidia CEO Jensen Huang told Bloomberg all of this is going to pass, though he sells the chips.
- Austin's take
- Austin agrees with Huang and calls it a blip and an opportunity. He explains that chips are physical infrastructure needing financing, so rising inflation and Treasury yields raise borrowing costs, and he flags circular financing, like Nvidia helping fund data centers, as a domino risk. Unlike the dot-com era, he argues, real demand exists out to 2029 or 2030, and older GPUs are staying useful longer than expected. He still expects consolidation and winners and losers.
- Why it matters
- If you own an S&P 500 fund, you own a big slice of these chip stocks.
- What to do
- Know your exposure through index funds before reacting to daily swings, and weigh the borrowing and inflation risks against the demand story.
Watch from 17:17
- What happened
- SoftBank is reportedly considering buying Allied Universal, America's largest security guard company, in a deal worth about $18 billion. The company has about 770,000 workers. Sources say SoftBank would eventually use robots for some guard tasks, with human-looking robot guards taking longer. Bloomberg says founder Masayoshi Son wants the deal to put robots to work. SoftBank already agreed to buy ABB's robot business for about $5.4 billion. SoftBank won't comment.
- Austin's take
- Austin sees it as the physical-world version of a private equity playbook: buy labor-heavy businesses and replace some work with AI to raise margins, like firms doing it with accounting. He likes robots taking on dangerous situations but finds protection by robots dicey, and says the job displacement question from the Anthropic story has to be solved.
- Why it matters
- It would be a test of robots at real workforce scale, far beyond household helpers, with hundreds of thousands of jobs in the picture.
- What to do
- Watch whether a deal actually happens; if you work in a labor-heavy field, notice that this buy-and-automate playbook is spreading.
- Not confirmed yet
- This is early and based on unnamed sources. Nobody has said how many jobs robots would take.
Watch from 18:18
- What happened
- The NSA put unexplained numbers into a common security standard in 2000. Claude worked out their origin in under thirty minutes: they came from computer clock times in 1997 and 1998, fed through an old random number generator. Cryptographer Filippo Valsorda, who ran the $36,864 bounty, redid the work himself and got the same numbers. The money goes to charity. Anthropic used an internal model not available to the public, and one of the numbers is still unexplained.
- The claim
- Anthropic says this isn't a hack and doesn't weaken the security standard.
- Austin's take
- Austin notes that cryptography was missing from OpenAI's recent math announcements, and says the bigger story is AI solving in thirty minutes what experts couldn't in 25 years. He calls the long-feared NSA backdoor apparently false and says the standard stays safe to use.
- Why it matters
- People feared for over two decades that these numbers hid an NSA backdoor in widely used security; that now looks much less likely.
- What to do
- No action needed on the standard itself. Do expect more long-standing technical puzzles to fall to AI quickly.
- Not confirmed yet
- One number remains a mystery, and the model used isn't public, so outsiders can't fully repeat the AI side of the work.
Watch from 24:18
- What happened
- The NBA told sportsbooks to stop offering some popular bets: who scores first, the opening tip, and bets that players who barely play will come up short. It built an AI that watches injury reports, social media, and betting patterns. This follows gambling cases tied to Terry Rozier, Chauncey Billups, and Jontay Porter. The league and sportsbooks have used tools like this for years, and the NBA still partners closely with sportsbooks.
- The claim
- The league says the tool will help it investigate unusual bets.
- Austin's take
- Austin calls it a great move to remove ways to cheat and wonders whether the AI surfaced insider activity or the league just saw those bets as too easy to rig. He sees sports as data science at its core and mentions his brother-in-law's work with a company that predicts baseball injuries to guide rest and rotation.
- Why it matters
- If you bet on the NBA, your options just shrank and the league is watching more closely.
- What to do
- If you wager, expect fewer prop bets and more scrutiny of unusual patterns.
- Not confirmed yet
- It's not clear how new the tool really is, since similar monitoring has existed for years.
