Research & Benchmarks - Page 20 of 28
Academic AI research, performance benchmarks, scientific breakthroughs, and peer-reviewed studies advancing artificial intelligence frontiers.
547 articles
View complete article list
Analysis overhauls AI Index; GPT-5.2 beats professionals on 70.9% of tasks
The AI industry has stopped bragging about test scores and started clocking in. A major research group has scrapped its old benchmarks for measuring intelligence. They now look at whether models can do a job, not pass a test.
MIT study probes memorization risk of clinical AI with de-identified data
We strip names from medical data and call it safe. A new MIT study says that safety is mostly an illusion. Clinical AI models, trained on mountains of supposedly anonymous patient records, can be made to cough up private details.
AMD announces Ryzen AI 400 at CES, resembles AI 300 in laptops
At CES, AMD announced a new laptop processor. It looks exactly like the old one. The chipmaker calls it the Ryzen AI 400 series, codenamed Gorgon Point. This is a technical truth.
Docker Trick: Deterministic OS Packages in One Layer to Prevent ML Failures
Your machine learning pipeline runs perfectly on your laptop. On the cluster, it crashes with a cryptic `libgomp` error.
Notion’s simplified AI agent feature feels indispensable, says engineer
Notion's latest AI trick is obvious. So obvious it took an engineer having a colleague literally tell him to talk to the machine like a person. Nystrom, a software engineer, builds things with strict logic. He's used to deterministic systems.
DeepSeek's architectural fix improves large-scale reasoning, follows GRPO work
DeepSeek appears profoundly uninterested in an arms race defined by spending. Its latest research instead drills deeper into a stubborn, simple conviction: the real key to advanced AI reasoning isn't just more computing power, but smarter...
Nested Learning's Continuum Memory System Redefines AI Memory for 2026
Most AI memory is a single, brittle block. Nested Learning is trying something else. The startup treats memory as a spectrum—a series of modules that update at wildly different frequencies.
New framework lets agentic AI tools adapt to fill main agent knowledge gaps
The most sophisticated AI agents still stumble, they lack specific knowledge, get stuck on niche tasks, or hemorrhage accuracy when pushed beyond their training data. That gap is where the real value lies.
Opera Neon: AI-native browser that researches, compares prices, codes
Every browser company now says AI is its copilot. Opera Neon suggests the AI should fly the plane and file the flight plan, too. This isn't about a smarter search bar. It's a different piece of software entirely.
Fusion reactors could produce dark-sector particles via neutron emissions
The promise of fusion power has long been tangled in a paradox: infinite clean energy, yet a stubbornly elusive reality. But what if these reactors could do more than light our cities?
CIOs drive AI experiments by embedding ready-to-use features into everyday tools
Forget the staged product launch. The real enterprise AI revolution is buried in the hum of your Monday morning: a suggested reply in Outlook, a formula debugged in Excel, a Jira ticket auto-sorted.
Dell and NVIDIA Host AI Developer Meetup in Hyderabad to Discuss Solutions
Every tech vendor is now throwing an AI mixer. Dell and NVIDIA are hosting one in Hyderabad on May 10, but they're bringing a specific prop: a developer machine you can touch. The event is a meetup, not a summit.
Google highlights AI-driven chip, infrastructure and robotics advances in 2025
Google's 2025 research review reads like an engineering manifesto. It is less about new chatbots and more about AI that builds things, controls things, and makes the systems we already depend on run with a new kind of efficiency.
LeCun and Hassabis dispute meaning of ‘general intelligence’
The debate over artificial general intelligence often feels like a battle over a mirage. Yann LeCun, Meta’s chief AI scientist, says the mirage is us.
Qwen3-4B-Instruct-2507: 4B-parameter model boosts Raspberry Pi AI
The Raspberry Pi used to be a joke for serious AI. You could run a model, sure, if you didn't mind the thing thinking for five minutes about lunch. That's over. Qwen3-4B-Instruct-2507 changes the rules.
Dell and NVIDIA host AI developer meetup in Bengaluru on deployment trade-offs
AI demos are pure fantasy. The hangover hits at deployment. On January 17, 2026, in Bengaluru, Dell and NVIDIA are convening a private session to confront that specific headache.
GPT-5.2 leads FrontierScience test, but falters on real research tasks
A new benchmark lands, and GPT-5.2 sits at the top. OpenAI's FrontierScience test is designed to be grueling: each problem demands three to five hours of work, scored on a ten-point rubric, with the model grading itself.
OpenUSD and NVIDIA Halos Enhance Robotaxi Safety with Synthetic Data, SimReady
The road to safe, autonomous mobility has never been more data-hungry. Robotaxis must navigate infinite edge cases, icy crosswalks, jaywalking pedestrians, the chaotic dance of a busy intersection.
Audio Dataset Valuable for Listening Models, Tackles Noise, Accents, Timing
The silence of a text dataset is deceptive. It never mumbles, never trips over a word, never battles the hum of an air conditioner in the background.
Fastweb and Vodafone use LangGraph LLM Compiler to automate customer requests
Most customer service AI is a scripted dead end. Fastweb and Vodafone built a bossy one that thinks. For 9.5 million of their Italian customers, a request doesn't trigger a canned reply.
Featured Resources & Reviews
No Code MBA Course Review
Learn to build AI-powered apps without coding. Our comprehensive review of No Code MBA's course.
AI Tools & Resources
Curated collection of AI tools, courses, and frameworks to accelerate your AI journey.
Weekly AI Digest
Get the week's most important AI news delivered to your inbox every week.