CharlieKey takeaways from your favorite podcasts
← 28 October 2026TBPN

Anduril’s American-Made Laptops, OpenAI Disproves ETH, New MacBook Pro | Diet TBPN

Anduril and Qualcomm's US-made Blue Chip laptop anchors a news roundup that also covers OpenAI's claimed disproof of the exponential time hypothesis, Anthropic's secret-successor structure, and DoorDash's Instacart talks.

Watch on YouTube ↗

Charlie's summary

  1. Anduril's Blue Chip laptop is a bet that a country that can't build its own computers can't secure them

    Anduril and Qualcomm announced Blue Chip, an American-made laptop due in 2027 and built for government, defense, first responders and public safety workers. In the launch video Palmer says “Any country that can't build its own computers can't secure them.” The hosts tie it to reindustrialization and to backdoor risk in foreign chips. They note it's the first US-made laptop in far longer than the 20 years Palmer claims, and they would like a consumer version.

    ▶ 1:01
  2. Blue Chip's launch video works because its B-roll shows only what is relevant

    The hosts contrast Anduril's video with copycat vibe reel launches inspired by the 2020 Founders Fund annual meeting. Here the footage matches the narration: wartime computing history while Palmer talks history, then the actual computer. They call that the way to do a launch video and say the vibe reel format should be shelved until about 2050, when it could come back as a nod to the 2020s.

    ▶ 7:08
  3. Apple's laptop design has stagnated, so a touchscreen MacBook looks incremental

    Walking through US laptop history from the Osborne 1 to the 1991 PowerBook, the hosts say design improved quickly and then stalled. The new 14-inch M6 MacBook Pro is thinner, lighter and faster but looks like the last one. They see the rumored touchscreen version as incremental, and they wonder whether Apple will reinforce the hinge. They also argue that the Jobs era had odd products too.

    ▶ 13:15
  4. OpenAI's claimed disproof of the exponential time hypothesis redraws a line computer scientists were confident about

    A post explains that the hypothesis held that no algorithm could solve 3-SAT without exponential blowup. An internal OpenAI model reportedly found a faster algorithm and checked the proof in Lean. The hypothesis's original authors and MIT's Ryan Williams are leading the public review. The algorithm is still slow and this isn't P versus NP, but the map of what's possible was wrong where confidence was high.

    ▶ 17:48

Give your email to unlock the rest of the key points — for every episode.

Do more with Charlie

Free — sign in with your email, no password.