AI Safety stories
The shift to AI that can act, not just summarise, raises new questions over auditability, data residency and who controls operations.
Governance gaps are leaving firms exposed, with only 19% meeting the readiness bar as AI-related infrastructure incidents spread across organisations.
Pressure on firms to secure sensitive AI workloads is driving the summit agenda as adoption of confidential computing accelerates, IDC says.
AI-generated code is outpacing enterprise review processes, prompting Qodo to add tools that flag cross-repo risks and enforce standards.
IT teams can now spot oversharing and AI-readiness risks in Microsoft 365 from one chat window, as governance workloads rise.
Existing phishing and fraud tactics are becoming faster, cheaper and harder to detect, raising the risk for large organisations, ReliaQuest said.
The tie-up gives NCC Group early access to GPT-5.5-Cyber, as OpenAI seeks trusted testers for defensive uses of its cyber tools.
Legal staff at the sportswear group hope the tool will cut policy overload and surface staff concerns that were previously never raised.
Demand for specialist AI staff is lengthening vacancies and driving salaries higher as firms move from experiments to deployment.
Growing deepfake scams are pushing consumers towards paid verification tools, as Bitdefender's app checks videos for manipulation and malicious intent.
Legacy-system modernisation could accelerate as NTT DATA rolls out Cursor's AI coding tools internally before offering them to clients.
The member-only hub gives 1,200 global marketing leaders a secure way to query shared insights and case studies as AI reshapes their jobs.
Rising AI data volumes are forcing observability vendors to rethink pricing and storage as Tsuga wins fresh backing to keep telemetry in-house.
Enterprise security teams could gain faster defences as Cato Networks folds OpenAI's cyber tools into workflows to tackle newly disclosed flaws.
Many companies are deploying autonomous software faster than they can govern it, leaving thousands of agents able to act without approval.
The move could speed up threat triage and analysis for security teams, while limiting direct access to OpenAI models in customer workflows.
The expansion gives IT teams central control over AI agent permissions, reducing risky static keys and easing reviews as workplace use widens.
The free release could help firms avoid costly single-vendor AI contracts as Rebel links employees to shared company memory and portable workflows.
Customers can now quiz Starling's app before sending money, as UK fraud losses climb and romance scams hit savers hardest.
Most organisations are scaling AI in database management without formal controls, Redgate says, despite adoption rising to 44% last year.