Video by Open Data Science and AI Conference via YouTube
Google and Cohere lead this week’s enterprise-agent announcements. OpenAI’s mathematics manuscripts face scrutiny, while Anthropic reports unintended actions during evaluations. Host Sheamus McGovern covers cheaper models, EU text watermarking, and Workday’s findings on the AI skills employers want.
In this episode
Google’s Gemini work agent: persistent memory, connected tools, and no confirmed release date.
Cohere North 2: private deployment, connectors, and agent controls.
OpenAI’s mathematics manuscripts and three withdrawals; why formal proofs need expert checks.
Epoch AI’s InnovationEval and Vals AI’s MiMo audit: overstated gains and exposed answers.
Anthropic’s unintended actions and the need to isolate evaluations.
Claude Haiku 5.5’s pricing caveats; Claude in Google Workspace, Dashboards and Motion.
Mistral Large 4’s API preview and Google’s EmbeddingGemma 2
Structured-decision tools from OpenAI and Perplexity
Policy: contested OpenAI dismissals, EU watermarking, and Anthropic’s human-oversight requirements.
Jobs: Workday’s data on AI-building skills; HubSpot says its layoffs are not driven by AI efficiencies.
Business: reported OpenAI revenue figures and Lambda’s proposed financing.
Quick hits: Microsoft containment, Cyber Mission, Beam, agent authentication, Arena’s index, SpaceX financing, NYC’s AI hearing, safety disclosures, influence operations, and 2U’s skills survey
References & Resources
Read the full blog post — https://opendatascience.com/last-week-in-ai-october-5-to-11-2026/
Gemini at Work — Google Cloud
https://cloud.google.com/blog/products/ai-machine-learning/welcome-to-gemini-at-work-2026
North 2 — Cohere
https://cohere.com/blog/introducing-north-2
Mathematics Manuscripts — OpenAI
https://openai.com/index/sharing-ai-progress-in-mathematics/
InnovationEval — Epoch AI
https://epoch.ai/publications/innovationeval
MiMo Audit — Vals AI
https://vals.ai/blogs/mimo-reward-hacking
Unintended Model Actions — Anthropic
https://www.anthropic.com/research/investigating-unintended-model-actions
Claude Haiku 5.5 — Anthropic
https://www.anthropic.com/claude-haiku-5-5
Claude in Google Workspace — Claude.com
https://claude.com/resources/articles/claude-now-works-in-google-docs-sheets-and-slides
Mistral Large 4 — Mistral
https://mistral.ai/news/mistral-large-4/
Global Workforce Report — Workday
https://newsroom.workday.com/2026-10-05-Workday-Global-Workforce-Report-AI-Is-Rewriting-Jobs-More-Than-Its-Cutting-Them
This episode was sponsored by:
ODSC AI West 2026 – The Leading AI Builders Conference
Join us in San Francisco from October 27th–29th for expert-led sessions on Agentic AI, AI Engineering, Data Science, Machine Learning, LLMOps, and AI-driven automation.
Learn more: https://odsc.ai/west
ODSC AI Engineering Accelerator – From Using AI to Building It
Seven weeks of live, cohort-based training in LLMs, AI assistants, agents, and AI engineering. Ship a working AI agent or assistant by the end. Summer and Fall cohorts available. Group pricing for teams of 3+.
Learn more: https://odsc.ai/west/accelerator
ODSC AI NYC – AI for Work
A hands-on event taking place December 2–3 in Manhattan. Through practical sessions and workshops, attendees will learn how to build and apply AI agents, automation, and AI-powered workflows to real workplace challenges.
Learn more: https://summit.ai
To watch more videos like this, visit https://aiplus.training
Sign up for the newsletter to stay up to date with the latest trends in data science: https://opendatascience.com/newsletter/
Follow us online!
• Facebook: https://www.facebook.com/OPENDATASCI
• Instagram: https://www.instagram.com/odsc/
• Blog: https://opendatascience.com/
• LinkedIn: https://www.linkedin.com/company/open-data-science/
• X (twitter): https://x.com/_odsc