Anthropic stories
Advertisers could shift more performance budgets to the open web as Taboola adds real-time budget allocation and AI campaign tools.
Boards face mounting pressure to fix AI-found code flaws faster, as CrowdStrike and partners launch a service to rank exploit risks.
As cyber tools become more powerful, Anthropic is limiting access while OpenAI is widening it, raising fresh fears over misuse.
Hundreds of packages could have exposed API keys and logins after Claude Code saved approved commands in a file npm may publish by default.
Customers will build and manage AI agents in one place as Google Cloud folds Vertex AI services into Gemini Enterprise Agent Platform.
AI-driven attacks are pushing firms to hide systems from the public internet rather than rely on patching flaws after discovery.
Marketing teams can now link Adobe tools with outside AI services under a governed system aimed at auditable customer experience workflows.
Production AI is straining as 5% of model requests fail and almost 60% of those errors stem from capacity limits.
Capacity limits are now behind nearly 60 per cent of AI production failures, risking outages and higher costs as usage scales.
Analysts can now triage threats and trace outages from inside AI tools, as Elastic’s public preview cuts dashboard switching.
AWS users can now run GitLab Duo Agent Platform inference through Amazon Bedrock, keeping AI workflows inside existing controls and budgets.
Boards are being pressed to abandon periodic patching as AI models can now uncover and chain software flaws faster than human teams can respond.
Financial regulators are alarmed after Anthropic said Claude Mythos can uncover software flaws at machine speed, raising bank security risks.
Users of Claude should see fewer peak-time slowdowns as Anthropic secures more AWS capacity and Amazon adds USD $5 billion to its stake.
Security teams are bracing for harder-to-stop attacks after the model found a Linux kernel flaw that had gone unnoticed for 27 years.
Customers across New Zealand and Australia can now get broader access to Claude models through Lancom, as AI projects shift from trials to live use.
The expansion will lift MongoDB's Irish headcount by more than 50% by 2027 as it adds engineering and AI roles in Dublin and Cork.
Security chiefs say unauthorised access to Anthropic AI's Mythos model shows generative tools could speed phishing, scanning and exploit discovery.
The tie-up could speed secure AI adoption for regulated Japanese firms, with NEC set to roll out Claude to about 30,000 staff.
Businesses in Southeast Asia can now access Google Cloud tools that connect AI agents, data and security, with chip and Workspace upgrades.
