AI Safety stories
The move aims to help defenders turn faster vulnerability discovery into working fixes, as OpenAI broadens access to its cyber tools and partners.
Non-lawyers in procurement, sales and operations can now use the platform, as the company seeks to cut routine work in legal departments.
The record outlay underscores the group's push into AI, health and payments as emissions fell and carbon neutrality continued for a fifth year.
Pre-orders open in the United States, United Kingdom and France for Snap's standalone glasses, priced at USD $2,195 and due later this year.
Mac users at many firms can now be covered by the same AI data-loss rules as Windows, closing a governance gap for sensitive work.
Private preview access is now available as security teams race to govern AI agents and harden identity controls for a post-quantum era.
Bad contact records can send autonomous AI workflows off course, with errors compounding across thousands of customer actions at once.
Most large US enterprises say AI agents are creating unmanaged financial and compliance risks, with many forced to reverse their actions.
The promotion puts KnowBe4's product strategy under an internal engineering veteran as the company expands defences against AI-driven threats and human error.
The open-source release gives enterprises a single control layer for fragmented AI agent tools, with governance and cost controls built in.
AWS customers building AI agents gain policy enforcement and recovery tools as Rubrik extends its governance layer into Bedrock AgentCore.
The tie-up adds tighter access checks as firms deploy AI agents and browser tools more widely, amid rising identity attacks.
Unsafe AI-generated clips could be stopped before playback under Ion's Australian patent filing, aimed at blocking deepfakes and other harmful video at source.
Skills shortages are leaving New Zealand firms exposed as AI adoption outpaces cyber and governance expertise across key sectors.
The expansion gives Korean firms, researchers and officials wider access to Claude as Anthropic deepens its push in one of Asia's busiest AI markets.
But 56 per cent of users rely on unapproved tools, leaving Australian employers to tackle security, compliance and trust gaps.
Rising AI failures are forcing firms to isolate response teams from the corporate network as incidents multiply across models and agents.
Boards may be missing who gains and who loses from AI, as a new paper says efficiency metrics can mask wider social and cultural costs.
Law firms face rising risk from fabricated case references as BriefCatch rolls out a tool to check citations before filings go in.
Financial firms face tighter oversight as the regulator warns current controls are not enough for fast-changing AI systems and machine identities.
