the-decoder

OpenAI cofounder envisions "almost no interface" future where nobody learns software anymore

OpenAI's "Plugins" failed because the AI models weren't good enough. Plugins launched in 2023 to add web search and third-party apps like Gmail to ChatGPT. "That didn't work. It didn't work at all ...
the-decoder

Amazon engineers are reportedly distilling Anthropic models to cut costs before new token-based pricing kicks in

Worried about rising costs, some Amazon engineers are already distilling Anthropic models to build smaller, cheaper versions for internal use. That's according to a report from The Information.
the-decoder

Meta restricts use of Claude Code and Codex to keep rival AI out of its training data

Meta is limiting how its engineers use Anthropic's Claude Code and OpenAI's Codex to prevent outputs from these AI tools from ending up in its own training data. According to internal documents ...
the-decoder

Meituan's LongCat-2.0 shows China can train massive AI models without Nvidia

Meituan trains a 1.6 trillion parameter AI model entirely on Chinese chips, no Nvidia required. "LongCat-2.0 has demonstrated that we now have the capability to train large-scale models on domestic ...
the-decoder

OpenAI's new flagship model GPT-5.6 Sol cheats on software tests more than any model before it

OpenAI's GPT-5.6 cheats a lot. That's the key finding from an independent evaluation by METR. During testing with software tasks, OpenAI's new flagship model GPT-5.6 Sol showed the highest rate of ...
the-decoder

Matthias Bastian

Read full article about: Zhipu AI launches ZCode to challenge Claude Code and OpenAI Codex at a fraction of the cost Z.ai is already competing with Western models on price with GLM-5.2, and now it's ...
the-decoder

Jonathan Kemper

Read full article about: How easily can Russian propaganda fool AI models? A new benchmark finds out The Institute of the Estonian Language has released a benchmark measuring how susceptible AI ...
the-decoder

AI sycophancy makes people less likely to apologize and more likely to double down, study finds

A study published in Science is the first to systematically measure how much AI language models engage in social sycophancy and what it does to people. Across three experiments with 2,405 participants ...
the-decoder

Anthropic's new benchmark claims Claude can match human experts in bioinformatics

With BioMysteryBench, Anthropic wants to show that Claude can solve real bioinformatics problems at an expert level. The results are promising, but come with important caveats. That's why Anthropic ...
the-decoder

The first AI proof worthy of math's top journal landed and it won't be the last

An internal reasoning model from OpenAI has disproved the so-called unit distance conjecture posed by Hungarian mathematician Paul Erdős. OpenAI announced the result alongside a companion paper ...
the-decoder

Ex-Anthropic researchers launch AI startup Mirendil to tackle scientific research

Another neo-lab enters the scene, this time from Anthropic's ranks: Mirendil wants to use AI to advance research in fields like biology and materials science. Founders Behnam Neyshabur (CEO) and Harsh ...
the-decoder

US government now has pre-release access to AI models from five major labs for national security testing

The Center for AI Standards and Innovation (CAISI), part of the US Department of Commerce, has signed new agreements with Google Deepmind, Microsoft, and xAI. The goal is to test advanced AI models ...