Category: AI & Tools

Claude Code Auto Mode Config: hard_deny, soft_deny, and allow Coffee Summary FACT (docs): Auto mode routes tool calls through a classifier that blocks irreversible, destructive, or out-of-environment actions; permissions.deny / permissions.ask still run before the classifier. FACT: Org trust is configured mainly via autoMode.environment (repos, buckets, domains, services) — destinations not listed are treated as…

Mistral’s $3.5B Raise: European Sovereign AI Meets Data-Center Ambition Coffee Summary FACT (CNBC, Sep 8, 2026): Mistral raised €3 billion (~$3.5 billion); post-money valuation more than €21 billion (~$24B in CNBC framing). FACT: Round led by Samsung Electronics; also involved EU-backed Scaleup Europe Fund (EQT-managed) and existing investor PSG Equity. FACT: Prior valuation cited: €11.7…

NVIDIA Buys Hugging Face for $12.93B: What Stays Open for Builders Coffee Summary FACT: NVIDIA announced it has agreed to acquire Hugging Face for $12,930,300,000 (blog dated September 3, 2026, by Jensen Huang). FACT (NVIDIA): Hugging Face will remain an open platform — builders keep model, framework, cloud, inference-provider, and compute choice; NVIDIA compute will…

Claude Managed Agents Permissions: always_allow vs always_ask Defaults Coffee Summary FACT: Permission policies govern server-executed tools only (agent toolset + MCP). Custom tools are app-side and not covered. FACT: Policies are always_allow, always_ask, and (as of 10 Sep 2026) auto — server evaluates each call to run, deny, or pause for approval. FACT: Defaults: agent…

Claude Smart Reports Beta: What Enterprise Admins Actually Get Coffee Summary FACT: Smart reports (beta) launched for Claude Enterprise on 10 Sep 2026 — analyze team usage: work done, cost, friction, and patterns worth packaging as shared skills. FACT: During beta, each org can run up to 10 reports per month free; the limit resets…

Pace the Frontier: Amodei Slowdown Call, Altman Backing, and the Safety-Body Talks Coffee Summary FACT: On 12 Sep 2026 Anthropic CEO Dario Amodei published "We Must Pace the Frontier," arguing labs should slow capability gains so safety work can keep up — not halt training. FACT: Anthropic unilaterally commits to "embedded evaluators" (third parties with…

Claude Smart Reports Beta: What Enterprise Admins Actually Get Coffee Summary FACT: Smart reports (beta) launched for Claude Enterprise on 10 Sep 2026 — analyze team usage: work done, cost, friction, and patterns worth packaging as shared skills. FACT: During beta, each org can run up to 10 reports per month free; the limit resets…

Pace the Frontier: Amodei Slowdown Call, Altman Backing, and the Safety-Body Talks Coffee Summary FACT: On 12 Sep 2026 Anthropic CEO Dario Amodei published “We Must Pace the Frontier,” arguing labs should slow capability gains so safety work can keep up — not halt training. FACT: Anthropic unilaterally commits to “embedded evaluators” (third parties with…

Claude Text Watermark and Detection API Access (EU AI Act) Coffee Summary FACT: Anthropic is rolling out SynthID-Text-style watermarking so Claude-generated text can be statistically checked for Claude involvement (EU AI Act / Code of Practice transparency). Watermarking changes the source of randomness in low-stakes token choices — no hidden characters, no per-user tracing, negligible…

GPT Image 2.5 Sunburst vs Flare: Which Model Should You Ship? Coffee Summary FACT (changelog Sep 8, 2026): OpenAI released GPT Image 2.5 Sunburst and GPT Image 2.5 Flare for Image API and Responses image generation. Decision rule: Sunburst when edit precision matters; Flare for fast everyday generation. Both support xhigh and max quality settings.…