| | OpenAI researcher announced GPT-5 math breakthrough that never happened (the-decoder.com) |
| 428 points by Topfi 6 days ago | past | 234 comments |
|
| | Google boasts 1.3 quadrillion tokens each month (the-decoder.com) |
| 1 point by jasoncartwright 14 days ago | past |
|
| | Anthropic bans companies majority-controlled by China, Russia, Iran, North Korea (the-decoder.com) |
| 8 points by amai 32 days ago | past | 1 comment |
|
| | Anthropic uses dark patterns to gain user consent for AI data use in Claude (the-decoder.com) |
| 3 points by alwillis 56 days ago | past |
|
| | Claude models can now end conversations with abusive users (the-decoder.com) |
| 2 points by alwillis 68 days ago | past |
|
| | U.S. authorities embedded secret GPS trackers in shipments of advanced AI chips (the-decoder.com) |
| 4 points by gmays 69 days ago | past | 2 comments |
|
| | OpenAI's o3 model bests the newer GPT-5 model on complex, multi-app office tasks (the-decoder.com) |
| 3 points by alwillis 69 days ago | past |
|
| | Bill Gates does not expect GPT-5 to be much better than GPT-4 (2023) (the-decoder.com) |
| 3 points by geox 71 days ago | past |
|
| | Anthropic CEO admits compromising with authoritarian regimes to get AI funding (the-decoder.com) |
| 10 points by alwillis 3 months ago | past | 4 comments |
|
| | Anthropic could soon be worth $100B – thanks to Claude Code (the-decoder.com) |
| 3 points by alwillis 3 months ago | past | 1 comment |
|
| | AI coding can make developers slower even if they feel faster (the-decoder.com) |
| 4 points by amai 3 months ago | past | 1 comment |
|
| | Claude now connects with Canvas, Panopto, and Wiley for educational access (the-decoder.com) |
| 2 points by alwillis 3 months ago | past |
|
| | Google unveils MedGemma, an open-source AI model suite for medical applications (the-decoder.com) |
| 1 point by alwillis 3 months ago | past |
|
| | Musk unveils Grok 4 as xAI's new AI model that beats OpenAI and Google (the-decoder.com) |
| 1 point by thm 3 months ago | past | 1 comment |
|
| | Amazon is producing a film about OpenAI's 2023 leadership crisis (the-decoder.com) |
| 1 point by geox 3 months ago | past |
|
| | Deepseek faces possible removal from German app stores (the-decoder.com) |
| 2 points by amai 4 months ago | past |
|
| | Chatbots have not led to significant changes in wages or working hours (the-decoder.com) |
| 3 points by geox 4 months ago | past |
|
| | New Othello experiment supports the world model hypothesis for LLMs (the-decoder.com) |
| 1 point by danielmorozoff 4 months ago | past |
|
| | Meta's latest model highlights the challenge AI faces in long-term planning (the-decoder.com) |
| 4 points by danielmorozoff 4 months ago | past |
|
| | Feeding AI models 10% 4chan trash actually makes them better behaved (the-decoder.com) |
| 2 points by Qadriq 4 months ago | past |
|
| | Apple study finds "a fundamental scaling limitation" in LLM reasoning models (the-decoder.com) |
| 9 points by rblion 4 months ago | past | 2 comments |
|
| | Reddit sues Anthropic for scraping site content to train Claude (the-decoder.com) |
| 4 points by 01-_- 4 months ago | past | 1 comment |
|
| | OpenAI Launches Infrastructure Partnerships for Countries (the-decoder.com) |
| 2 points by deadstack 5 months ago | past |
|
| | Seedream 3.0 beats GPT-4os image generation, says ByteDance (the-decoder.com) |
| 3 points by t0lo 6 months ago | past |
|
| | ChatGPT reaches new growth milestones and runs into capacity issues (the-decoder.com) |
| 2 points by speckx 6 months ago | past |
|
| | Deepseek reportedly restricts employee travel amid AI security concerns (the-decoder.com) |
| 12 points by qwertox 7 months ago | past |
|
| | Claude Jailbreak results are in, and the hackers won (the-decoder.com) |
| 4 points by dlikren 8 months ago | past |
|
| | OpenAI funded independent math benchmark before setting record with o3 (the-decoder.com) |
| 56 points by rar00 9 months ago | past | 6 comments |
|
| | [dupe] Eliza, world's first chatbot, back online after 60 years thanks to AI historians (the-decoder.com) |
| 4 points by bookofjoe 9 months ago | past | 1 comment |
|
| | OpenAI's O1-preview model manipulates game files to force a win in chess (the-decoder.com) |
| 5 points by nocoder 9 months ago | past | 1 comment |
|
|
| More |
