Deepseek stories
The free release could help firms avoid costly single-vendor AI contracts as Rebel links employees to shared company memory and portable workflows.
The grant lets the London startup train an air-gapped coding model on UK infrastructure, bolstering supply for defence and other sensitive sectors.
Editors facing EU transparency rules can now flag AI-made text, images and audio before publication, with fines possible for non-compliance.
The new tool aims to catch Bitcoin software flaws between formal audits after a regression led to more than USD $116 million stolen.
Up to 4.16 times faster vector ingestion could help OpenSearch customers cut AI search latency and reduce storage overhead.
Developers can now access a smaller Russian-made model aimed at coding and long-document analysis, as Sber opens GigaChat 3.5 Ultra free.
Four of six major AI sales agents posted negative scores in a new benchmark, highlighting the cost of poor prospecting output for businesses.
Enterprises face higher AI bills and governance gaps as only 17 per cent have reached high maturity, Gartner says.
Agentic AI, zero-day surge, sovereign cloud, and humanoid robots will define IT strategy in 2027, Info-Tech Research Group warns.
Companies may be exposing sensitive data as staff use personal AI accounts for work nearly two-thirds of the time, researchers found.
Most workers are blurring the line between corporate and personal AI use, leaving employers blind to sensitive data shared outside approved accounts.
Workplace AI use is rising faster than company oversight, with a small minority of staff driving most activity and security risks.
Australian startups will get direct access to Chinese tech giants, with a Zhejiang trade mission including Alibaba, Unitree Robotics and Geely.
Australian businesses can now access locally run AI inference as SCX.ai charges by token and sidesteps hyperscale training costs.
AI projects that remain stuck in pilot stages may miss the shift to agentic systems able to manage tasks with limited human oversight.
Patients seeking jobs or care may face bias from chatbots, with a new study finding health stigma can skew AI outputs in subtle ways.
Americans largely misunderstand how chatbots source brand answers, leaving companies at risk of being invisible in AI search results.
Production users can now route generative AI requests through a stable open source gateway, with Bloomberg already running it and Nutanix adopting it.
Voluntary model reviews may leave gaps as advanced AI systems move closer to critical infrastructure and enterprise data.
Current frontier models still fall short of stand-alone cyber defence, with the top performer spotting only 46% of attack evidence in Simbian’s test.