The Brief

New briefs daily around 7 AM Eastern

Guardrail every MCP tool call natively in OpenAI's Agents SDK

OpenAI's Agents SDK v0.22.1 adds built-in guardrails that block MCP tool calls before they execute, plus configurable sandbox isolation, so agent builders can enforce a safety policy without writing custom code around every tool call.

• 01Read

  SemiAnalysis benchmarks show TPUv7 beats Nvidia on cost

  SemiAnalysis's first third-party benchmarks put Google's TPUv7 Ironwood up to 50% ahead of Nvidia's B200/B300 on inference performance per dollar, based on a direct comparison of internal and external TCO.

  Alec Ibarra→
• 02Read

  Uber's DragonCrawl lets LLMs test mobile apps like a human

  Uber's DragonCrawl has LLMs read mobile app screens and decide what to tap the way a human tester would, adapting to UI changes across 50+ languages without script maintenance.

  Uber Engineering 1→
• 03Listen

  Roman Ugarte details building Grok Bot in seven weeks

  Lenny's Newsletter interviews the xAI engineer who took Grok Bot from first commit to public launch in about seven weeks, including why the team built it standalone instead of inside Cursor.

  Lenny Rachitsky→
• 04Read

  DeepMind maps every possible human DNA point mutation

  Google DeepMind's AlphaGenome Atlas precomputes the regulatory impact of all 9 billion possible single-nucleotide variants in the human genome as a queryable, 1-petabyte dataset.

  blog.google→
• 05Read

  OpenAI's Pachocki defends racing toward defensive AI

  OpenAI chief scientist Jakub Pachocki argues the strongest case for continued fast capability scaling is building defensive AI to secure infrastructure and counter rogue agents in real time.

  simonwillison.net→