The BaseOfficialAnthropic
Introducing the Anthropic Cyber Mission
Anthropic announces the Anthropic Cyber Mission, which it calls a long-term commitment to support defenders. It launches a Critical Infrastructure Defense Program, giving security providers for power, water and transport systems Claude models, on-site engineers and threat research, and OSS Scanner, free model-generated vulnerability scans for open-source projects, sent without human review.
Why it matters
The same thing, for different readers.
What it means for you
Anthropic says the cost of exploiting software flaws has dropped, and is offering its models to people who defend power grids, water systems and widely used free software.
For lawmakers
Anthropic says it has offered models and support to more than half of US states; officials may ask who oversees private AI access to critical infrastructure.
For researchers
Anthropic expects above 90% true positives from unreviewed, model-generated bug reports and forecasts AI favoring defense in two years; both are claims others can check.
The question it raises
When the same models can help attackers and defenders, who should decide which defenders get access, and on what terms?
In their own words
Every quotation was found, word for word, in the text we saved from the source.
From the source’s own text.
“As part of Project Glasswing, partners uncovered many vulnerabilities, but we haven’t yet achieved a sufficient reduction in cyber risk.”
“Highly cyber-capable AI models are widely available to attackers now. But defensive tools—including our own—have not yet reached enough of the defenders who need them.”
“The reports are model-generated and sent without human review. That means maintainers receive them faster, but it also means that some will contain inaccuracies, such as a wrong severity rating.”
“Our forecast is that in two years, AI will favor defense: it will be easier to catch bugs before they ship, write fundamentally secure software from scratch, and actively defend systems with models. But in the near term, that may not be true.”
The record
- Questions
- How is AI used in attacks, defence and war? · Are AI companies putting stronger safeguards into practice?
- People
- Tony Baker (Speaks, VP, Digital Trust & Chief Product Security Officer, Rockwell Automation) · Harshul Joshi (Speaks, Partner, US OT/ICS Security Lead, Cyber Data and Tech Risk, PwC) · Sam Rubin (Speaks, Senior Vice President, Unit 42, Palo Alto Networks) · Moreno Carullo (Speaks, Co-Founder and CTO, Nozomi Networks) · Dan Gunter (Speaks, CEO and Founder, Insane Cyber) · Jon Lavender (Speaks, Chief Technology Officer, Dragos) · Daniel Bernard (Speaks, Chief Business Officer, CrowdStrike) · Hiroaki Koiwa (Speaks, Head of Cyber CoE, AI & Software Services Business Unit, Hitachi, Ltd.) · Adnan Amjad (Speaks, US Cyber Leader, Deloitte & Touche LLP) · Andrew Turner (Speaks, President of Commercial Cyber, Booz Allen) · Harpreet Sidhu (Speaks, Global Cybersecurity Lead, Accenture)
- What we read
- We read the whole source.
- Found by
- The Radar, run of October 12, 2026, via
anthropic-news(tier 1: the source’s own channel or an official record) - Cite this
- Anthropic, “Introducing the Anthropic Cyber Mission”, October 8, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/anthropic-2026-10-08-cyber-mission