Meta Pauses Work With Mercor After Data Breach Puts AI Industry Secrets at Risk
Major AI labs are investigating a security incident that impacted Mercor, a leading data vendor. The incident could have exposed key data about how they train AI models.
Living brain cells enable machine learning computations
A research team at Tohoku University and Future University Hakodate has demonstrated that living biological neurons can be trained to perform a supervised temporal pattern learning task previously carried out by artificial systems. By integrating cultured neuronal networks into a machine learning framework, the team showed that these biological systems can generate complex time-series signals, …
Read more “Living brain cells enable machine learning computations”
LTX Desktop 1.0.3 is live! Now runs on 16 GB VRAM machines
The biggest change: we integrated model layer streaming across all local inference pipelines, cutting peak VRAM usage enough to run on 16 GB VRAM machines. This has been one of the most requested changes since launch, and it’s live now. What else is in 1.0.3: Video Editor performance: Smooth playback and responsiveness even in heavy …
Read more “LTX Desktop 1.0.3 is live! Now runs on 16 GB VRAM machines”
Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment
Despite their sophisticated general-purpose capabilities, Large Language Models (LLMs) often fail to align with diverse individual preferences because standard post-training methods, like Reinforcement Learning with Human Feedback (RLHF), optimize for a single, global objective. While Group Relative Policy Optimization (GRPO) is a widely adopted on-policy reinforcement learning framework, its group-based normalization implicitly assumes that all …
Read more “Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment”
Smarter Live Streaming at Scale: Rolling Out VBR for All Netflix Live Events
By Renata Teixeira, Zhi Li, Reenal Mahajan, and Wei Wei On January 26, 2026, we flipped an important switch for Live at Netflix: all Live events are now encoded using VBR (Variable Bitrate) instead of CBR (Constant Bitrate). It sounds like a small configuration change, but it required us to revisit some of the foundational assumptions …
Read more “Smarter Live Streaming at Scale: Rolling Out VBR for All Netflix Live Events”
Simulate realistic users to evaluate multi-turn AI agents in Strands Evals
Evaluating single-turn agent interactions follows a pattern that most teams understand well. You provide an input, collect the output, and judge the result. Frameworks like Strands Evaluation SDK make this process systematic through evaluators that assess helpfulness, faithfulness, and tool usage. In a previous blog post, we covered how to build comprehensive evaluation suites for …
Read more “Simulate realistic users to evaluate multi-turn AI agents in Strands Evals”
How Honeylove boosts product quality and service efficiency with BigQuery
Building the perfect bra takes thousands of data points. That’s why Honeylove isn’t just another intimates brand. We’re a technology company that happens to make exceptional bras, tops, shapewear, and bodysuits. Technology shapes everything we do, from how we iterate garments based on customer feedback to how we optimize sizing across those thousands of data …
Read more “How Honeylove boosts product quality and service efficiency with BigQuery”
‘Uncanny Valley’: Iran’s Threats on US Tech, Trump’s Plans for Midterms, and Polymarket’s Pop-up Flop
In this episode, we discuss Iran’s threats to target US tech firms, gear up for the midterm elections, and get a scene report from the Polymarket pop-up bar in DC.
Crashing waves vs. rising tides: Overturning prior views about how AI could overtake human workers
Anthropic CEO Dario Amodei has said that AI could surpass “almost all humans at almost everything” shortly after 2027. While AI’s capabilities are certainly improving, such rapid progress might seem at odds with findings that show AI is still failing at 95%+ of remote freelance projects, and continues to struggle with hallucination, long term planning, …
Comparing 7 different image models
Tested a couple of prompts on different models. Only the base model, no community-made loras or finetunes except for SDXL. I’m on 8gb of vram so I used GGUFs for some of these models which is likely to have diminished the results. My results and observations will also be biased just from my personal experience, …