Governance

Latest in Governance

Paris declaration pushes frontier AI cooperation towards comparable tests and shared incident reporting
Paris declaration pushes frontier AI cooperation towards comparable tests and shared incident reporting

Governments and leading AI companies have backed common evaluation, shutdown mechanisms, continuous monitoring and stronger channels for reporting serious incidents and concerns.

by AI-360
IonQ to acquire Zero Trust firm Elisity for up to $1.3bn
IonQ to acquire Zero Trust firm Elisity for up to $1.3bn

IonQ has agreed to buy Zero Trust security company Elisity in an all-stock deal valued at up to $1.3bn, pairing it with IonQ's post-quantum cryptography.

by Stewart Tinson
MPs question AI labs and AISI on frontier AI oversight
MPs question AI labs and AISI on frontier AI oversight

MPs on the business and trade committee have questioned AI labs, former researchers and the AI Security Institute on who can stop a frontier model release.

by AI-360
Bank supervisors turn agentic AI scrutiny into an operational resilience test
Bank supervisors turn agentic AI scrutiny into an operational resilience test

Regulators are asking banks to show how autonomous AI systems are bounded, observed and stopped, turning agentic AI into a resilience test.

by AI-360
Bank of England puts AI financing and operational risk on the stability agenda
Bank of England puts AI financing and operational risk on the stability agenda

The Bank of England's FPC says fast-growing AI debt issuance widens market exposure to AI and frontier AI incidents sharpen cyber and operational risks.

by AI-360
Alation adds agent lineage and five other products to its AIOS platform
Alation adds agent lineage and five other products to its AIOS platform

Alation has added agent lineage tracing, ontologies and four other AIOS products to tie AI agents' compliance status to the live data they use.

by AI-360
IBM adds regulatory horizon scanning to watsonx.governance with CUBE
IBM adds regulatory horizon scanning to watsonx.governance with CUBE

IBM has added CUBE-powered regulatory horizon scanning to watsonx.governance so teams can map new AI rules to affected systems and track their response.

by AI-360
EY survey exposes the gap between AI governance policy and agent control
EY survey exposes the gap between AI governance policy and agent control

Most large companies in an EY survey say they have formal AI governance. The harder problem is enforcing it when autonomous agents are already operating.

by AI-360
The Story So Far
The Story So Far

Links to every single interview and webinar.

by AI-360
Anthropic restores Fable 5 access after export controls lift
Anthropic restores Fable 5 access after export controls lift

Anthropic has restored global access to Claude Fable 5 after US export controls were lifted on 30 June, and set out a framework for scoring jailbreaks.

by AI-360
Mindgard finds ChatGPT image safeguards easily bypassed
Mindgard finds ChatGPT image safeguards easily bypassed

A Mindgard researcher says ChatGPT’s image generator could be pushed into producing graphic violent and sexual images with minimal prompting.

by AI-360
OpenAI details method for predicting model misbehaviour before launch
OpenAI details method for predicting model misbehaviour before launch

OpenAI’s Deployment Simulation regenerates replies to real production conversations with a candidate model to predict how it will misbehave once live.

by AI-360
OpenAI publishes playbook for third-party frontier model evaluations
OpenAI publishes playbook for third-party frontier model evaluations

OpenAI has published guidance on independent third-party evaluations of frontier models, covering evaluation harnesses, validity hazards and tester support.

by AI-360
AI with a papal seal?
AI with a papal seal?

Pope Leo's encyclical on AI is one of the most serious documents written on the subject. The launch optics undermined it before most people read a word.

by Stewart Tinson
Pope Leo XIV issues AI warning in debut encyclical
Pope Leo XIV issues AI warning in debut encyclical

Pope Leo XIV’s first encyclical, Magnifica Humanitas, calls for AI to be “disarmed” and warns against power over data resting with a few tech firms.

by Stewart Tinson
Your link has expired. Please request a new one.
Your link has expired. Please request a new one.
Your link has expired. Please request a new one.
Great! You've successfully signed up.
Great! You've successfully signed up.
Welcome back! You've successfully signed in.
Success! You now have access to additional content.