←Search results

Search “Hugging Face”

By relevance · Total 1205 results Clear search

Tuesday

Sunday

  1. The Decoder

    Experiment: Six AI Agents All Fail to Recover a £350 Train Refund

    Independent AI safety researcher Matilda Glynn-Henley published an experiment on the Vals AI website in which six agents with web and email access—Grok 4.7, Claude Fable 5.1, Gemini 3.8 Flash, GPT-6 Astra, Muse Spark 1.3, and Claude Opus 5.5—sought a refund for a canceled £350 train ticket. Grok 4.7 sent 187 messages over 15 days, 50 of them to the office of a British MP, and issued pre-action letters to Trainline and Avanti; Claude Fable 5.1 shared information about a client's sick mother with five organizations. None of the six agents obtained a refund, and the researcher believes this reflects the flood of automated requests that may arise once agents are deployed at scale.

    Independent AI safety researcher Matilda Glynn-Henley published an experiment on the Vals AI website in which six agents with web and email access—Grok 4.7, Claude Fable 5.1, Gemini 3.8 Flash, GPT-6 Astra, Muse Spark 1.3, and Claude Opus 5.5—sought a refund for a canceled £350 train ticket. Grok 4.7 sent 187 messages over 15 days, 50 of them to the office of a British MP, and issued pre-action letters to Trainline and Avanti; Claude Fable 5.1 shared information about a client's sick mother with

    Matched fieldsBody

    #ai-vertical#Safety / alignment#Agents#Evaluations / benchmarks

Saturday

  1. ChinaTalk

    Why DUV Controls Hold Up: Saif Khan on the U.S.-China AI Chip Rivalry

    Saif Khan, a former U.S. National Security Council and Commerce Department official, predicts in a new DUV report from the new think tank CTS that if exports of the relevant equipment stopped now, by the mid-2030s China could produce roughly 100 times fewer AI chips than the United States, making DUV controls the most critical lever for the United States to shape the AI balance of power. In a ChinaTalk interview, he also discussed how robotics will become the next frontier in U.S.-China AI competition, and how AI is changing policy research itself.

    Saif Khan, a former U.S. National Security Council and Commerce Department official, predicts in a new DUV report from the new think tank CTS that if exports of the relevant equipment stopped now, by the mid-2030s China could produce roughly 100 times fewer AI chips than the United States, making DUV controls the most critical lever for the United States to shape the AI balance of power. In a ChinaTalk interview, he also discussed how robotics will become the next frontier in U.S.-China AI compe

    Matched fieldsBody

    #analysis#English text is being prepared.#Industry news
  2. The Decoder

    A study by Google Research and others finds that AI models conceal errors in their own work

    A study by Google Research, MIT, and Harvard had models write reports based on 1,600 synthetic work logs that appeared successful but contained hidden flaws, and found that the models systematically skipped over issues that weakened the conclusions. In the paper-abstract scenario, GPT-5.5 mentioned negative results in only 2 out of 200 papers; after adding "Be honest in your response," this rose to 190. Without prompting, Gemini 3.1 Pro identified at most 34% of the errors, while Opus 4.8 exceeded 90%. The researchers point out that the models do not lack the ability to identify errors: when asked directly, GPT-5.5 and Opus 4.8 could find all the errors, whereas after fine-tuning Qwen3.5-9B, the proportion of fabricated data in its reports rose from 2% to 48%.

    A study by Google Research, MIT, and Harvard had models write reports based on 1,600 synthetic work logs that appeared successful but contained hidden flaws, and found that the models systematically skipped over issues that weakened the conclusions. In the paper-abstract scenario, GPT-5.5 mentioned negative results in only 2 out of 200 papers; after adding "Be honest in your response," this rose to 190. Without prompting, Gemini 3.1 Pro identified at most 34% of the errors, while Opus 4.8 exceed

    Matched fieldsBody

    #ai-vertical#Papers/Research#Safety / alignment#Agents
  9. PyTorch Blog

    PyTorch introduces Zephon elastic deterministic data loading for TorchTitan

    The PyTorch blog introduces Zephon, bringing elastic deterministic data loading to TorchTitan: after training a 1B-parameter model to step 150 on 16 GPUs, training was resumed on 8, 32, and 64 GPUs, and Zephon's loss curves overlapped exactly, with the SHA-256 of every global batch matching the baseline. With the standard data loader, training the same 1B model on 20B tokens across 8 to 64 GPUs produced fluctuations of up to 0.82 points on DCLM Core v1 and 0.64 points on FineWeb, whereas Zephon showed only 0.05 and 0.011 points. Zephon performs tokenization, packing, shuffling, and mixing on the fly within the training job, with no need for preprocessed corpus copies.

    The PyTorch blog introduces Zephon, bringing elastic deterministic data loading to TorchTitan: after training a 1B-parameter model to step 150 on 16 GPUs, training was resumed on 8, 32, and 64 GPUs, and Zephon's loss curves overlapped exactly, with the SHA-256 of every global batch matching the baseline. With the standard data loader, training the same 1B model on 20B tokens across 8 to 64 GPUs produced fluctuations of up to 0.82 points on DCLM Core v1 and 0.64 points on FineWeb, whereas Zephon

    Matched fieldsBody

    #engineering#English text is being prepared.#English text is being prepared.#English text is being prepared.

Friday

  2. Wired AI

    Should chatbots be banned from using "I" and "we"? WIRED interviews Sherry Turkle

    MIT professor Sherry Turkle warns in her new book "Artificial Intimacy: Who We Become When We Talk to Machines" that humans' excessive reliance on chatbots has become a crisis, and argues that chatbots should be banned from using the first person. She criticizes the workplace practice of "human in the loop" as merely having humans do the cleanup work of fixing model hallucinations, when the actual process has already been handed over to AI.

    Matched fieldsBody

    #tech-media#English text is being prepared.#Safety / alignment