AI Safety stories
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.
The funding values the cybersecurity group at USD $12 billion as enterprises race to secure data exposed to AI tools and agents.
Manufacturers risk compliance failures and production delays if they treat AI-generated code as a shortcut to rebuilding core ERP systems.
The move aims to help Wipro turn AI pilots into client workflows, as it trains 10,000 staff to deploy Claude across industries.
The move puts Broadridge among firms using frontier AI to harden financial software, where breaches can disrupt trading and client communications.
Weaker oversight could turn AI-generated code into a costly drag, with security flaws and technical debt rising in enterprise projects.
Enterprise teams using AI coding tools may face higher technical debt, security gaps and costs, according to new SIG research.
Security and governance tools are being added as enterprises push agentic AI from pilots into live production systems.
HPE has expanded its self-driving networking strategy with new AI, security and data centre capabilities across enterprise environments.
The integrations aim to close security gaps as more firms run AI in production across gateways, APIs and models.
Enterprise teams are getting a single control plane to track agent sprawl, tighten permissions and curb AI spending as autonomous systems spread.
Financial firms face tighter oversight as the regulator warns current controls are not enough for fast-changing AI systems and machine identities.
Android users will get a warning when a supposedly familiar caller may be spoofed, as deepfake-enabled scams rise.
Enterprises can now turn plain-language requests into reviewable AI workflows, as Dataiku seeks to close the gap between prototypes and production.
The lender expects AI to speed fraud checks and staff support, while helping prioritise projects that could each deliver more than USD $100 million.
Growing fears over deepfakes have pushed Ion to seek Australian patent protection for a method that verifies video at the byte level.
