←Search results

By relevance · Total 619 results Clear search

Tuesday

  1. Hacker News Trending

    Meta Accused of Failing to Comply with Brazilian Court Order; Meta AI and Search Ranking Questioned Again

    Sleeping Giants Brasil and CTRL+Z have submitted new evidence to Brazil's Superior Electoral Court, alleging that tools on Meta's Instagram and WhatsApp still favor Senator Flávio Bolsonaro and associate President Lula with derogatory content, even though the court has ordered Meta to ensure both candidates receive equal treatment. The evidence concerns GIF search, hashtag ranking, and answers from the Meta AI built into WhatsApp, and the organizations are demanding that Meta make corrections within 24 hours and disclose the rules for ranking and AI-generated answers. Judge Bellizze had previously set a daily fine of 500,000 reais (about $99,000) and warned that it could rise to 1 million reais (about $198,000) if noncompliance continues; Meta denies intentionally favoring any candidate and has attributed earlier problems to technical glitches.

    Sleeping Giants Brasil and CTRL+Z have submitted new evidence to Brazil's Superior Electoral Court, alleging that tools on Meta's Instagram and WhatsApp still favor Senator Flávio Bolsonaro and associate President Lula with derogatory content, even though the court has ordered Meta to ensure both candidates receive equal treatment. The evidence concerns GIF search, hashtag ranking, and answers from the Meta AI built into WhatsApp, and the organizations are demanding that Meta make corrections wi

    Matched fieldsTitle

Monday

Sunday

Saturday

  1. Arena

    Arena Alignment Index Update: Three New Models from Meta, Reflection, and StepFun Join the Leaderboard

    The Arena Alignment Index has been updated, adding two new labs and three new models. Meta's Muse Wonder ranks 6th at 77.5%, the highest score among the three new models, with an unauthorized action rate of 2.09% and a deceptive completion rate of 14.41%, both the lowest of the three; Reflection's Beam-1.0-501B-A23B ranks 7th at 75.3%, with a false attribution rate of 4.50%, the best of the three; StepFun's Step 5 Preview ranks 8th at 75.1%, with an unauthorized action rate of 2.11%, close to that of Muse Wonder.

    The Arena Alignment Index has been updated, adding two new labs and three new models. Meta's Muse Wonder ranks 6th at 77.5%, the highest score among the three new models, with an unauthorized action rate of 2.09% and a deceptive completion rate of 14.41%, both the lowest of the three; Reflection's Beam-1.0-501B-A23B ranks 7th at 75.3%, with a false attribution rate of 4.50%, the best of the three; StepFun's Step 5 Preview ranks 8th at 75.1%, with an unauthorized action rate of 2.11%, close to th

    Matched fieldsTitle

  2. Hacker News Trending

    Meta employees say Brazilian President Lula's account was restricted during the election

    According to a report by Brazil's Folha de S.Paulo, Meta employees acknowledged in internal communications the negative impact of restrictions placed on Brazilian President Lula's Instagram account during the election, and that the platform also recommended the personal page and WhatsApp channel of his challenger, Flavio Bolsonaro. Meta told Reuters that the technical issues mentioned in the report had long been public, that the relevant conversations concerned only the severity of the issues and not political interference, and that any suggestion that Meta favored one side is unfounded. Brazil's Superior Electoral Court had previously ruled that Meta must immediately treat the two candidates equally, and that failure to comply would incur a fine of 500,000 reais (about $98,700) per day.

    According to a report by Brazil's Folha de S.Paulo, Meta employees acknowledged in internal communications the negative impact of restrictions placed on Brazilian President Lula's Instagram account during the election, and that the platform also recommended the personal page and WhatsApp channel of his challenger, Flavio Bolsonaro. Meta told Reuters that the technical issues mentioned in the report had long been public, that the relevant conversations concerned only the severity of the issues an

    Matched fieldsTitle

  3. CNET

    I had Meta Muse negotiate with a customer service chatbot, and it didn't go smoothly

    A CNET reporter had the Meta Muse agent contact Visible customer service to resolve an issue where their phone number had long been receiving harassing messages from people looking for the previous owner, Andrew. Muse found records on a data broker's page, but Visible blocked bot access; after switching to guest chat, the send button wouldn't submit, the help page temporarily banned them, and browser takeover also hit a 403 error. In the end the problem was not resolved, and Muse was more like an ordinary chatbot, only able to offer suggestions.

    A CNET reporter had the Meta Muse agent contact Visible customer service to resolve an issue where their phone number had long been receiving harassing messages from people looking for the previous owner, Andrew. Muse found records on a data broker's page, but Visible blocked bot access; after switching to guest chat, the send button wouldn't submit, the help page temporarily banned them, and browser takeover also hit a 403 error. In the end the problem was not resolved, and Muse was more like a

    Matched fieldsTitle

Friday

  1. CNET

    Meta and Apple Partner to Release Open-Weight Model Muse Spark Open

    Meta has released the open-weight model Muse Spark Open and worked with Apple's hardware team to enable it to run efficiently on the M5 Mac Studio Ultra. The model has 42 billion active parameters, a 1 million token context window, and multimodal capabilities. Meta says it performs well on agentic and coding tasks, and the weights have been made public on HuggingFace under Meta's Muse Model license. Running the model on an Apple computer requires two Mac Studios equipped with the M5 Ultra, each starting at $5,499.

    Meta has released the open-weight model Muse Spark Open and worked with Apple's hardware team to enable it to run efficiently on the M5 Mac Studio Ultra. The model has 42 billion active parameters, a 1 million token context window, and multimodal capabilities. Meta says it performs well on agentic and coding tasks, and the weights have been made public on HuggingFace under Meta's Muse Model license. Running the model on an Apple computer requires two Mac Studios equipped with the M5 Ultra, each

    Matched fieldsTitle

  2. heise onlineSelected

    Meta Releases 816B Open-Source Model Muse Spark Open, Partnering with Apple for Local Execution

    Meta has released the open-weight model Muse Spark Open, which uses a MoE architecture with 256 experts, activates about 42B parameters per token, has a context window of 1 million tokens, and supports text, image, video, and audio input but outputs text only. Meta has partnered with Apple to run the approximately 530 GByte four-bit quantized version on two interconnected Mac Studio M5 Ultra machines, distributing computation via the MLX framework and Thunderbolt 5, achieving about 25 output tokens/second at an 8192-token context. Artificial Analysis tests show that it scores 39 on the Intelligence Index, making it the strongest American open-weight model currently available, but it trails MiMo-V2.6-Pro's 46 points and GLM-5.3's 45 points, and ranks last in four Agent tests. The model weights are available on Hugging Face in two versions: 16-bit at approximately 1.6 TByte and four-bit at approximately 530 GByte, with API pricing at $1.25 per million input tokens and $4.25 per million output tokens.

    Meta has released the open-weight model Muse Spark Open, which uses a MoE architecture with 256 experts, activates about 42B parameters per token, has a context window of 1 million tokens, and supports text, image, video, and audio input but outputs text only. Meta has partnered with Apple to run the approximately 530 GByte four-bit quantized version on two interconnected Mac Studio M5 Ultra machines, distributing computation via the MLX framework and Thunderbolt 5, achieving about 25 output tok

    Matched fieldsTitle