Microsoft's new AI training method eliminates bloated system prompts without sacrificing model performance

In building LLM applications, enterprises often have to create very long system prompts to adjust the model’s behavior for their applications. These prompts contain company knowledge, preferences, a [...]

Why LinkedIn says prompting was a non-starter — and small models was the breakthrough

LinkedIn is a leader in AI recommender systems, having developed them over the last 15-plus years. But getting to a next-gen recommendation stack for the job-seekers of tomorrow required a whole new [...]

MIT's new fine-tuning method lets LLMs learn new skills without losing old ones

When enterprises fine-tune LLMs for new tasks, they risk breaking everything the models already know. This forces companies to maintain separate models for every skill.Researchers at MIT, the Improbab [...]

Researchers baked 3x inference speedups directly into LLM weights — without speculative decoding

As agentic AI workflows multiply the cost and latency of long reasoning chains, a team from the University of Maryland, Lawrence Livermore National Labs, Columbia University and TogetherAI has found a [...]

How I Get Free Traffic from ChatGPT in 2025 (AIO vs SEO)

Three weeks ago, I tested something that completely changed how I think about organic traffic. I opened ChatGPT and asked a simple question: "What's the best course on building SaaS with Wor [...]

Lego's latest educational kit seeks to teach AI as part of computer science, not to build a chatbot

Last week at CES, Lego introduced its new Smart Play system, with a tech-packed Smart Brick that can recognize and interact with sets and minifigures. It was unexpected and delightful to see Lego come [...]

Inside LinkedIn’s generative AI cookbook: How it scaled people search to 1.3 billion users

LinkedIn is launching its new AI-powered people search this week, after what seems like a very long wait for what should have been a natural offering for generative AI.It comes a full three years afte [...]

Nvidia's Nemotron-Cascade 2 wins math and coding gold medals with 3B active parameters — and its post-training recipe is now open-source

The prevailing assumption in AI development has been straightforward: larger models trained on more data produce better results. Nvidia's latest release directly challenges that size assumption â [...]

Anthropic says DeepSeek, Moonshot, and MiniMax used 24,000 fake accounts to rip off Claude

Anthropic dropped a bombshell on the artificial intelligence industry Monday, publicly accusing three prominent Chinese AI laboratories — DeepSeek, Moonshot AI, and MiniMax — of orchestrating coor [...]