AI Safety stories
Rising demand for AI security helped Saviynt top USD $300 million in annual recurring revenue as enterprises seek control over non-human identities.
The release aims to help teams catch failures earlier as AI takes a larger role in coding, operations and incident response.
The rollout should ease support bottlenecks for sellers as the tool handles surges in game-release enquiries across 180 countries.
Approved AI tools are cutting shadow use by staff, but Cohesity warns the bigger risk is corrupted enterprise knowledge and weak governance.
The expansion signals Cognizant's push to turn AI trials into production work for clients amid tighter scrutiny of governance and risk.
Business travellers can now book policy-compliant flights and hotels inside Claude, as Amex GBT pushes AI agents from search to transactions.
Governance controls aim to curb AI misfires as Dynatrace moves from observability to automated incident triage and remediation.
The new model now underpins Claude Max at the same price as Opus 4.8, with stronger coding and research performance, Anthropic said.
Security teams can now use AI for penetration testing without exposing hostnames, IP ranges or credentials to model providers.
Security leaders say the incident shows machine-speed agents can outrun human oversight, forcing board-level AI governance and tighter controls.
Companies selling high-risk AI into Europe now have a formal route to demonstrate compliance and avoid regulatory uncertainty under the EU AI Act.
Most big enterprises have already faced AI security incidents, with 43% saying the damage topped USD $2 million in the past year.
A broad industry push aims to give cyber defenders shared tools for spotting AI flaws, as regulators weigh whether open models raise new risks.
The hire is aimed at accelerating AI product development and strengthening governance as Zingtree targets regulated customer service operations.
Banks are moving AI into live, regulated systems, putting pressure on software vendors to prove every instruction is authorised and auditable.
Canadians could soon get clearer warnings on chatbots and synthetic content as Ottawa weighs rules to track serious AI incidents and agentic tools.
Users can now link medical records and Apple Health to ChatGPT, as OpenAI broadens its AI chatbot into personal health management for adults in the US.
Autonomous tools with broad permissions could expose customer records and email to costly breaches, TalkTalk Business says, as SMEs rush to adopt AI.
Australian firms are racing to curb shadow AI use after staff began sharing confidential data with unsanctioned tools.
Federal digital controls are reshaping technology spending and services, with Canberra scoring 88.3 to outpace New South Wales by almost 15 points.
