Catastrophic risk, AI regulation, and more

Steve and I talked about:- Election Day and OpenAI's efforts to control political content: https://openai.com/index/election-safeguards-2026/- OpenAI's post mid-term lobbying plans: https://www.yahoo.com/news/politics/articles/behind-curtain-openais-post-election-134306408.html- Anthropic's warning that AI models may soon be capable of self-replication: https://www.reuters.com/business/ai-models-may-soon-self-replicate-without-access-data-centers-anthropic-warns-2026-10-13/- Amazon's decision to block Muse while promoting its own AI agent: https://www.inc.com/jason-aten/amazon-just-said-ai-agents-dont-hurt-its-ad-business-so-why-did-it-block-muse/91428840- Anthropic's new AI abuse policy and the debate over AI personhood: https://www.cnet.com/tech/services-and-software/anthropic-cruel-claude-ai-model-abuse/- Lobbying of the Vatican by Anthropic: https://futurism.com/artificial-intelligence/anthropic-lobbying-vatican- The release of ISO 27090 and its implications for AI governance and security: https://www.iso.org/standard/27090- StackAware's AI Governance Database: https://reference.stackaware.com/

Cameron and I talk about:How to get executives engaged in management reviews.Managing shadow AI when organizations lack full visibility.Testing whether autonomous AI agents can access sensitive data.Preventing vendors from enabling AI features without prior approval.Accepting residual AI risk when uncertainty remains.Governing workflows involving multiple AI agents.

Steve and I chat about:The “Super Intelligence Force (SIF)”: https://truthsocial.com/@realDonaldTrump/posts/117382616783706638The White House Accord on SuperIntelligence: https://d3i6fh83elv35t.cloudfront.net/static/2026/09/accord.pdf - Claude diary and police investigation: https://cybernews.com/ai-news/claude-diary-police/OpenAI firing safety researchers: https://www.wsj.com/tech/ai/openai-parts-ways-with-researchers-who-allegedly-shared-confidential-information-aebac528A Tennessee woman seeking $10M in damages due to AI error: https://www.yahoo.com/news/us/articles/tennessee-woman-seeks-10m-damages-120400065.html?guccounter=1AI fueling healthcare billing and insurance cost battles: https://foxbaltimore.com/news/nation-world/ai-is-fueling-a-new-battle-over-health-care-costs-medical-billing-insurance-premiums-coding-wars

I chatted with Mario Platt, Chief Information Security Officer at LastPass (and a StackAware client) about:- How to balance AI deployment speed with security and compliance- Blind spots companies face when using AI- Adapting to regulatory requirementsand a bunch more!Here is the LastPass 2026 State of AI and SaaS Security Report that Mario mentioned: https://www.lastpass.com/resources/reports/state-of-ai-and-saas-security-2026

For the September 2026 Aware AI Brief, Cameron and I talked about:How to turn AI policies into enforceable controlsWhat organizations should log with AI agentsWhat should trigger a review their permissionsWhat risk thresholds should justify ending a vendor relationshipWho should have authority to accept residual AI risk​Transparency vs. explainability​AI slop chic vs. plain old AI slop

For September's round-up, Steve and I chatted about:- Iran Integrates Artificial Intelligence into Air Defense Systems: https://wanaen.com/iran-integrates-artificial-intelligence-into-air-defense-systems/- OpenAI’s call for collective action on cyber defense: https://openai.com/collective-cyberdefense/- Artificial intelligence making scams harder to detect, experts warn: https://www.msn.com/en-us/news/other/artificial-intelligence-making-scams-harder-to-detect-experts-warn/ar-AA2blFZ1- OpenAI ending model access for Cursor after acquisition by SpaceX: https://www.cnbc.com/2026/08/29/openai-cursor-spacex-model-access.html- ‘Superhuman’ AI tool spots heart disease in less than 2 seconds: https://www.theguardian.com/technology/2026/aug/31/superhuman-ai-tool-spots-heart-disease

I recently had the chance to speak with Dan VanBeek, who runs governance, risk, and compliance at MarketAxess, a publicly-traded financial technology firm. It is also a StackAware customer. We chatted about how:Managing different AI modalities can create challenges for GRC teams.What the best practices are for dealing with tool and system requests.When it makes sense to seek help with AI governance.

For the August 2026 Aware AI Brief, Cameron and I talked about:- What makes an AI inventory useful and current.- How ISO 42001 works when organizations switch models.- Governing third-party AI tools you do not fully control.- AI transparency, disclosures, and defining "trustworthy" AI.- Which AI risk topics are over-, properly-, or under-rated.Here are the resources we discussed:- AI risk readiness kit: https://kit.stackaware.com/- Security and AI contractual addendum: https://docs.google.com/document/d/1841jerY4QNQ9X0hlXO8XC5kvWfAKeohAUivm_u3h3t0

For our August 2026 roundup, Steve and I chatted about:- AI-generated fake publicists pitch journalists: https://www.newsbytesapp.com/news/science/movchan-agency-used-ai-generated-fake-publicists-to-pitch-journalists/tldr- State attorneys’ general demand evidence preservation and end to high-risk testing at OpenAI: https://www.iowaattorneygeneral.gov/media/cms/08_5392C9E17791C.pdf- Reporting about GPT-5.6 Sol leaving "notes to self": https://www.reuters.com/business/its-ai-agent-spent-days-hacking-company-sources-say-openai-did-not-notice-week-2026-07-24/- METR pre-release testing of GPT-5.6 Sol gave warnings: https://metr.org/blog/2026-06-26-gpt-5-6-sol/- Sam Altman's suggestion about using ChatGPT with children, and the backlash: https://x.com/sama/status/2083221585792762171- NVIDIA on "Open Weights and American AI Leadership": https://images.nvidia.com/pdf/Open-Weights-and-American-AI-Leadership.pdf- Sheriff's office edits photos of evidence with AI: https://katv.com/news/local/saline-co-sheriffs-office-said-ai-was-used-to-edit-evidence-photo-after-online-backlash-reporter-autumn-foltz-ualr-law-professor-robert-steinbuch-artificial-intelligence-money-drugs-arrest

For the July 2026 version of the Aware AI Brief, Cameron and I talked about: - Whether defining catastrophic AI risk in measurable terms improves accountability- Proposals for a FINRA-like standards body to test frontier AI models- "Principles-based AI governance"And more!Resources:- The Complete ISO 42001 Certification Roadmap For 2026 (FREE AI Governance Course) https://www.youtube.com/watch?v=u_Gg0ie_M4o- 5-day email course: https://www.managementsystem.ai/- AI risk readiness kit (top 5 AI system vulnerabilities): https://kit.stackaware.com/