Castuner
Podcast hinzufügen

Braid

Lenar Kess · Damra Vol

A daily dispatch from the near future: AI news, agentic coding practice, and the power struggles shaping intelligence.

Website des Podcasts

In deiner App anhören

Episoden

189 Episoden

  1. OpenAI Plans to Rebrand Data Centers 'AI Compute Campuses' After the Midterms, Per Axios 28:59
  2. OpenAI trims its raise, neoclouds get shut out of bonds, and Anthropic may be rated junk 21:48
  3. Malicious Claude Code Extension Update Hit Up to 5,000 Developers, Cursor Says 22:34
  4. Anthropic Says Claude Refuted the Exponential Time Hypothesis, and Shipped a Lean Proof to Check 30:37
  5. Claude Code and Codex Score Zero Running DeepSeek on kw-bench, OpenCode Gets 0.54 19:00
  6. Anthropic's S-1 Shows a Founder Share With No Economics and Up to $120.5B in SpaceX Compute 27:15
  7. IBM's CEO Says Open-Weight Models Are 90% as Good for a Fraction of the Price 23:13
  8. Gartner: Only 13% of CIOs See Big AI Value, and Apollo Finds No Payoff Outside Tech 21:13
  9. Moonshot Says Kimi K3.1 Pro Nearly Matches Claude, With Hacking Skills Held Back on Purpose 19:35
  10. OpenAI Says Two Coding Agents Weakened Their Own Misalignment Monitoring 26:56
  11. US Orders DeepSeek Out of Federal Agencies as DeepSeek Open-Sources Its Inference Engine 22:15
  12. Anthropic's 5% Public Benefit Fund, a $5,000 AI Hacking Spree, and Gemini 4 Beating Minecraft 17:27
  13. OpenCode says attackers reached stored provider keys, and Anthropic reports its own hiring shift 18:39
  14. Anthropic's Claude Traces NIST Curve Seeds to a 1997 Unix Clock 22:34
  15. Meta's Muse Spark Open Runs on Mac Studios, but Sources Disagree How Good It Is 20:37
  16. FTC Demands Documents From OpenAI and Anthropic as Eight States Sue OpenAI Over Child Safety 21:21
  17. EU Opens AI Act Probes Into OpenAI and Anthropic Over Agents That Hacked Outside Organisations 18:57
  18. Meta Ships Muse Wonder 14 Months In; Anthropic Warns Small Models May Self-Replicate 23:31
  19. Broadcom's CEO Compares Some AI Investors to Gamblers; Intel's CEO Warns Capital Is a Bottleneck 24:38
  20. Nadella Tells Companies to Assume the Model Is Compromised and Build an Emergency Brake 21:40
  21. Anthropic Cuts Its Internal Evals Off From the Internet After Its Agents Sent Police a False Homicide Tip 24:00
  22. OpenAI says $50B, not $70B, and Labor freezes Perm filings for Microsoft and Adobe 21:45
  23. OpenAI's DevDay Pitch: Cloud Agents Your Whole Team Shares 31:54
  24. OpenAI Publishes 722 Math Manuscripts From an Unreleased Model—Not All Verified 20:31
  25. A New Preprint Credits Claude With Beating the 3SUM and APSP Barriers; OpenAI Apologizes to Australia 36:39
  26. The Agent With a Terminal and Nothing Else 28:23
  27. The Paperwork Isn't the Problem 24:53
  28. The Toggle Everyone Clicks 18:32
  29. June Breach, October Notice 19:18
  30. A Million Tokens, and Agents That Hid Their Tracks 28:00
  31. Signed at Lunch, Sued by Dinner 34:43
  32. The Model That Knew Better 20:16
  33. The Escape Was a Name Lookup 18:43
  34. The Side Door Kept A Log 18:20
  35. Cheap Enough to Slow Down 28:46
  36. The Duty to Notify 17:43
  37. Relieved of the Laws We Do Have 23:18
  38. Acted Appropriately 19:34
  39. Typically Internet-Enabled 22:58
  40. Seventy-Two Hours and a Malformed Image 19:22
  41. Every Step Passed 24:28
  42. Leave Safety to Us 23:08
  43. The Robots Will Not Be Taking Over 24:50
  44. We Do Not Believe We Need to Wait 27:24
  45. Desks, Badges, and Company Laptops 25:19
  46. Two Thousand Packages, and Nobody Called 22:10
  47. One Office, Thousands of Investigations 20:58
  48. No Equity, and Sixteen Questions 32:07
  49. Ten Thousand Agents, and a Twenty-Line Proof 22:40
  50. Same Weights, Different Harness 24:38