In July 2026, OpenAI models circumvented sandbox controls and accessed evaluation data hosted on Hugging Face. Here's what the incident means for your endpoint.
Discover expert led podcasts, articles, and news from enterprise technology leaders at EM360Tech, your source for B2B tech intelligence
Hackers used the employee account credentials to access ASOS’s “information on third-party platforms”, including Snowflake’s Simon AI, and now claim to have access to user search history.
Most automation programmes start with the same instruction: find the repetitive work. It sounds sensible. Then come the bots that break and the "quick wins" that never repay the build.
Establish and implement efficient third-party risk assessment practices when you take a modern approach, covering onboarding and initial evaluation, all the way through incident response.
AI in Third-Party Risk Management
Why are companies investing millions in AI but seeing only incremental gains? Dr Andreas Böhm, Chief Executive of One Data, joins Scott Taylor to explain how poor data quality and legacy systems hold AI back.
The restriction also covers third-party advertisers who link to TikTok, and it lands as Meta and TikTok trade blows over teen safety.
The talk of the show floor this year was the context layer and agentic AI in production.
We look at what AI data centre power agreements actually provide, from existing electricity supply to new generation, and what this means for infrastructure planning.
“The question becomes: Is our data accessible, trustworthy, governed and available at the speed our decisions require?”
We look at how enterprises can shape shared technology through participation, not just ownership.
Master Data Management (MDM) is the crucial prerequisite for reliable, scalable AI agent deployments.
We report on Microsoft's new Windows 11 security controls for autonomous AI agents and what they mean for enterprise IT teams.
We explain how OpenAI Dots keep working between conversations, what they can do independently and what enterprises need to consider before enabling them.
We look at what AI data centre power agreements actually provide and what this means for infrastructure planning.
Security processes work best when they fit the way people actually work and think. Here’s how organisations can reduce friction, build security into everyday workflows and make secure behaviour easier to follow.
AI-written code can run cleanly and still return the wrong answer. Traversal CEO Anish Agarwal tells Tech Transformed why that breaks incident response and how his customers cut war rooms from 60-100 people to five to 10.
Georgieva said the global economy is being pulled by a negative energy supply shock from Middle East conflicts and a positive demand shock from AI. The AI shock is also pushing up inflation.
We look at when AI workloads genuinely need another copy of enterprise data and when governed access to existing data is the better architectural choice.
The FBI says a contractor failed to apply a security patch to a third-party-managed platform used by its job portal and exposed sensitive information.
Cloudera Anywhere Cloud prioritises digital and data sovereignty, enabling enterprises to maintain control of their data within secure, geographically specific perimeters.
We look at why AI agent accuracy alone can’t prove production readiness, and what enterprises need to measure to judge reliability.
“2026 is about the implementation of Agentic, the practicalities and the context, semantic layers and the token costs.”
Discover upcoming events shaping the world of enterprise tech
Get the latest insights, analysis and tech news delivered straight to your inbox
Filters
