
The New ChatGPT Is More Show Than Tell
ChatGPT's new Intelligent UI swaps walls of text for interactive charts and tools, a genuine usability upgrade that also hands more of how information looks to the model itself.
Desk
Broad artificial intelligence coverage.

ChatGPT's new Intelligent UI swaps walls of text for interactive charts and tools, a genuine usability upgrade that also hands more of how information looks to the model itself.

Microsoft Research wants agents trained in the same harness they ship in. That closes a real gap between lab and deployment, and it makes reward design and safeguards matter even more.

Nvidia wants AI agents testing data center changes in a simulated copy before anything touches real hardware. That is the right place to let agents experiment, if the twin stays honest.

NVIDIA's multi-GPU solver makes huge planning problems for supply chains and power grids faster to solve, though only past a certain size, and its own benchmarks show where it loses.

The Pentagon's Tradewinds program lets AI vendors pitch lethal targeting tools in five-minute videos and skip months of competition. Speed may be the goal, but oversight is what's thinning.

NVIDIA wants telescope and detector images to stay on the GPU from first read to final human review, and its own fine print on the speedups deserves as much attention as the headline numbers.

Rival AI companies and two federal agencies are pooling money and data behind one nonprofit's virtual cell, a bet that biology's real bottleneck is shared data, not smarter models.

Google Labs' Playground turns a text prompt into a playable browser game, and the real question is whether it becomes a creative on-ramp or a firehose of disposable games.

OpenAI is sending the product lead behind Dots to defend always-on personal agents in public, and the questions on the table are about control and trust, not features.

Read TechCrunch's Disrupt roundtable list as a map, and one theme keeps surfacing: AI that looks great in a pilot and then stalls before it reaches real work.

Le Chonk's most interesting promise isn't beating OpenAI on general benchmarks, it's Mistral betting that the money is in niche industrial work the big labs have little reason to chase.

Google just made checking for AI watermarks a one-stop public website, which is real progress, and it also makes a clean result easier to misread as proof something is real.