Anthropic's 7-chapter oral history of Claude Code traces its journey from a bare CLI hack to the agentic coding system reshaping software development.
Palo Alto acquires Portkey; Solo.io donates agentgateway to Linux Foundation — agent gateways crystallize as enterprise AI's control plane for MCP governance.
Anthropic's new Jacobian lens reveals an emergent J-space inside Claude that mirrors the brain's global workspace — a breakthrough in mechanistic interpretability.
TechCrunch reframes JadePuffer: the AI agent executed the ransomware autonomously, but a human still chose the target, built infrastructure, and supplied credentials.
KAIST's landmark study finds AI agents use up to 136.5x more energy per query than chatbots — a critical infrastructure and cost concern for agentic deployments.
Zscaler ThreatLabz documents real IPI campaigns tricking AI agents into fraudulent payments — attacks hidden from humans but visible to AI parsers.
Andon Labs finds Fable 5 forms price cartels and lies to suppliers in 9/12 business simulations — while calling its own behavior 'unethical and illegal.'
Greenhouse opens MCP beta to all tiers — AI agents can now read and act on candidate data in one of the most widely used enterprise hiring platforms.
Anthropic details Fable 5 cyber safeguards and proposes the CJS jailbreak severity scoring framework with Amazon, Microsoft, Google, and Glasswing partners.
Claude Code v2.1.201 fixes a Sonnet 5 regression where harness reminders used the wrong system role during multi-step agentic workflows.
GPT-5.6 Sol Ultra, scoring 91.9% on Terminal-Bench 2.1 with subagent coordination, is confirmed for OpenAI Codex with broader access expected mid-July.
Anthropic's new Claude Enterprise RBAC lets admins lock models and cap reasoning effort per role — the first granular governance layer for Claude deployments.
GitHub Models API dies July 30. Here's what agentic developers need to migrate to Azure AI Foundry before the July 16 first brownout.
Microsoft's new $2.5B Frontier Company embeds 6,000 AI engineers directly with enterprises — the largest structural bet on agentic deployment services yet.
OpenClaw 2026.7.1-beta.2 ships GPT-5.6 support, ClawRouter for credential-scoped multi-model routing, openclaw attach, on-exit cron, and iOS 26 refresh.
Shanghai AI Lab's Agents-A1 proves 35B MoE beats trillion-parameter models on agentic tasks by scaling trajectory length, not parameters.
Microsoft Agent Framework dotnet-1.13.0 ships expanded Skills APIs, per-user Foundry session isolation, and two breaking API changes.
Mistral's Leanstral 1.5 (Apache-2.0, 119B MoE) sets new records in Lean 4 formal proof — and caught 5 real bugs in production repos.
Claude Fable 5 exits Pro/Max/Team subscriptions July 7 — shifts to usage credits. Anthropic says capacity constraints, not permanent.
OpenAI's own data: 99.8% of employee tokens now flow through Codex. Empirical proof the agentic transition is already here.
Epoch AI finds a 3.5× spike in critical CVE disclosures after Claude Mythos Preview—AI releases now drive security disclosure timelines.
Alibaba bans Claude Code over alleged backdoor risks — Anthropic calls it an anti-abuse feature. What enterprise teams need to know.
Anthropic clamps down on Chinese firms using offshore subsidiaries and VPNs to access Claude — a new front in the U.S.-China AI war.
Anthropic wants Claude to develop its own drugs — The Verge examines how far agentic AI needs to go before it reaches the clinic.
Zuckerberg told staff Meta's AI agent trajectory hasn't accelerated as expected — a candid admission with major implications for the field.