Meta announced the release of two new frontier artificial intelligence models Monday as the technology company seeks to compete with similar products from Anthropic and OpenAI.
The two new models, Muse Wonder and Muse Wonder Full, are Meta’s “most capable models to date,” according to the company. Muse Spark, released in April, was the first AI model in the Muse family.
“Whether the job is navigating an insurance claim, understanding a lease, or making sense of someone’s finances, a model has to interpret the information in front of it and work through the problem,” Meta said in a press release. “Muse Wonder and Muse Wonder Full bring that capability to expert domains including law, finance, and health, along with frontier coding performance for people building everything from a game to an app for a new business.”
Muse Wonder and Muse Wonder Full will be equipped with two new capabilities that the company calls “Active Memory” and “Active Focus.” Active Memory refers to the ability to retain information from past conversations and recall details relevant to whatever task the user requests, and Active Focus allows the AI models to try different approaches and adapt to new information when obstacles arise.
Muse Wonder is already available through the Meta Model API and Muse Code, and the more powerful Muse Wonder Full model is only available for Muse Code subscribers.
Meta says the models are the next step toward giving users “personal superintelligence,” a term coined by Meta CEO Mark Zuckerberg to describe an AI companion that understands a person’s goals and preferences.
BIG TECH IS LEADING THE ‘AI SAFETY’ GUARDRAILS PUSH: FOR PROTECTION OR PROFIT?
The announcement comes amid heightened concerns over the rapid progress of frontier AI development, which lawmakers across various levels of government are trying to regulate in the name of safety.
Meta says it is also concerned about AI safety, which is why the company has taken steps to strengthen safeguards against cyber and biological misuse and reinforce protections for personal information. Meta will be releasing a report on safety evaluations soon.
Continue reading...
[ H/T Washington Examiner ]
