Category: AI Category


  • Mistral just updated its open source Small model from 3.1 to 3.2: here’s why

    Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more French AI darling Mistral is keeping the new releases coming this summer. Just days after announcing its own domestic AI-optimized cloud service Mistral Compute, the well-funded company has released an update to…

  • Hospital cyber attacks cost $600K/hour. Here’s how AI is changing the math

    Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more In years past, medical facilities weren’t as vulnerable as they are now; hackers had an unwritten rule not to target institutions or services where a disruption could put people in physical danger.…

  • Anthropic study: Leading AI models show up to 96% blackmail rate against executives

    Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Researchers at Anthropic have uncovered a disturbing pattern of behavior in artificial intelligence systems: models from every major provider—including OpenAI, Google, Meta, and others — demonstrated a willingness to actively sabotage their…

  • Google’s Gemini transparency cut leaves enterprise developers ‘debugging blind’

    Join the event trusted by enterprise leaders for nearly two decades. VB Transform brings together the people building real enterprise AI strategy. Learn more Google‘s recent decision to hide the raw reasoning tokens of its flagship model, Gemini 2.5 Pro, has sparked a fierce backlash from developers who have been relying on that transparency to build…

  • Unlock the other 99% of your data – now ready for AI

    For decades, companies of all sizes have recognized that the data available to them holds significant value, for improving user and customer experiences and for developing strategic plans based on empirical evidence. As AI becomes increasingly accessible and practical for real-world business applications, the potential value of available data has grown exponentially. Successfully adopting AI…

  • This post is divided into five parts; they are: • Understanding Positional Encodings • Sinusoidal Positional Encodings • Learned Positional Encodings • Rotary Positional Encodings (RoPE) • Relative Positional Encodings Consider these two sentences: “The fox jumps over the dog” and “The dog jumps over the fox”. Source link

  • Machine learning workflows often involve a delicate balance: you want models that perform exceptionally well, but you also need to understand and explain their predictions. Source link

  • Meta buys stake in Scale AI, raising antitrust concerns

    Meta’s $14.8 billion investment in Scale AI – and the hiring of the startup’s CEO – is drawing attention to how US regulators will handle acquihire-style deals under the Trump administration. The deal gives Meta a 49% nonvoting stake in Scale AI, which hires gig workers to label training data for AI systems. Scale’s clients…

  • This post is divided into three parts; they are: • Interpolation and Extrapolation in Sinusoidal Encodings and RoPE • Interpolation in Learned Encodings • YaRN for Larger Context Window Sinusoidal encodings excel at extrapolation due to their use of continuous functions: $$ \begin{aligned} PE(p, 2i) &= \sin\left(\frac{p}{10000^{2i/d}}\right) \\ PE(p, 2i+1) &= \cos\left(\frac{p}{10000^{2i/d}}\right) \end{aligned} $$ You…