data
Addressing hardware failures and silent data corruption in AI chips
Meta trained one of its AI models, called Llama 3, in 2024 and published the results in a widely covered paper. During a 54-day period of pre-training, Llama 3 experienced 466 job interruptions, 419 of which were unexpected. Upon further investigation, Meta learned 78% of those hiccups were caused by hardware issues such as GPU…
Is Your Data Storage Strategy AI-Ready?
The adoption of AI has caused an increased need for proper data governance, and companies are now under pressure to ensure data maturity. Globally, many companies are either using or exploring AI, with over 82% actively leveraging or considering AI for business operations. Yet, according to Gartner only 14% of cyber leaders can balance maximizing…
The quiet data breach hiding in AI workflows – Help Net Security
As AI becomes embedded in daily business workflows, the risk of data exposure increases. Prompt leaks are not rare exceptions. They are a natural outcome of how employees use large language models. CISOs cannot treat this as a secondary concern. To reduce risk, security leaders should focus on policy, visibility, and culture. Set clear rules…
New method efficiently safeguards sensitive AI training data
Data privacy comes with a cost. There are security techniques that protect sensitive user data, like customer addresses, from attackers who may attempt to extract them from AI models — but they often make those models less accurate. MIT researchers recently developed a framework, based on a new privacy metric called PAC Privacy, that could…
Salesforce AI Released APIGen-MT and xLAM-2-fc-r Model Series: Advancing Multi-Turn Agent Training with Verified Data Pipelines and Scalable LLM Architectures
AI agents quickly become core components in handling complex human interactions, particularly in business environments where conversations span multiple turns and involve task execution, information extraction, and adherence to specific procedural rules. Unlike traditional chatbots that handle single-turn questions, these agents must hold context over several dialogue exchanges while integrating external data and tool usage….
DeepSeek jolts AI industry: Why AI’s next leap may not come from more data, but more compute at inference
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More The AI landscape continues to evolve at a rapid pace, with recent developments challenging established paradigms. Early in 2025, Chinese AI lab DeepSeek unveiled a new model that sent shockwaves through the AI industry and resulted…
Is ChatGPT Stealing Your Face Data With Ghibli Trend? We Found Out
Hence, there is always a small possibility that OpenAI might be secretly storing your data somewhere and somehow, but there’s no concrete proof of the same, and this would be a baseless allegation at the moment. Nobody knows what happens to the data that you enter in ChatGPT, and it is unclear whether OpenAI stores…
Russian spy infiltrates ASML and NXP to steal technical data necessary to build 28nm-capable fabs
German A., a 43-year-old Russian engineer, is accused of secretly supplying sensitive technical information from ASML, NXP, and TSMC to Russia, allegedly to assist in building a 28nm-capable fab there, reports NRC. His illicit earnings were about €40,000, and he now faces 18 to 32 months in prison. Though German A. alone could not steal…
The TAO of data: How Databricks is optimizing AI LLM fine-tuning without data labels
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More AI models perform only as well as the data used to train or fine-tune them. Labeled data has been a foundational element of machine learning (ML) and generative AI for much of their history. Labeled data…
Making a case for the cybersecurity data fabric | TechTarget
Information, data and context are the weapons that cybersecurity teams use to battle adversaries daily. Yet, using cybersecurity data in the modern enterprise has become increasingly difficult. The data is often scattered among dozens of point technologies, fragmented with use-case-specific interfaces, and siloed between IT, security and application teams with little capability to support…