# AI news, Monday, 5 October 2026

Compiled by Richard Brooks's AI news desk. Canonical: https://richard-brooks.com/ai-news/2026-10-05/

## DeepSeek Harness adds experimental Claude Code Mods bridge

Tools. Open-source agent harness v0.2.1-alpha.1 treats Mods as a plugin subset. Compatibility is experimental, not a full port.

Source: DeepSeek, https://github.com/deepseek-ai/deepseek-harness/releases/tag/dsh-v0.2.1-alpha.1

## Microsoft and Hugging Face release ThinkingBox agent outcome benchmark

Agents. Grades agents on final database state across 507 workflows, each run 20 times. Most clean-looking failures still leave the wrong records.

Source: Hugging Face, https://huggingface.co/blog/microsoft/thinkingbox

## OpenAI safety leader David Robinson quits, cites broken culture

Business. He led launch safety reports and says trial-and-error deployment guarantees growing failures as agents get more capable. External incentives for lab safety are now part of the debate after rogue-agent incidents.

Source: TechCrunch, https://techcrunch.com/2026/10/03/openai-safety-employee-resigns-claiming-the-companys-culture-is-broken/

## Trump names DNI Jay Clayton as AI czar for Super Intelligence Force

Policy. A 120-day White House task force will report on AI risks and opportunities and recommend the federal role in oversight. Follows the already-filed voluntary White House accord with major labs.

Source: CNBC, https://www.cnbc.com/2026/10/03/trump-jay-clayton-ai-czar.html