SynopsisThe 29-year-old founder of data-labeling startup Scale AI joined Meta after Chief Executive Officer Mark Zuckerberg overhauled his company’s AI efforts, including through a $14 billion investment in Scale that was widely viewed as a bid to recruit Wang.ReutersAlexandr WangMeta Platforms Inc. Chief AI Officer Alexandr Wang said the company’s future artificial intelligence models will differentiate themselves from competitors through their consumer health capabilities. “Health is an area that we view as really critical as we scale these models out to billions,” Wang, who has helmed Meta’s artificial-intelligence strategy for the last year, said Thursday at the Bloomberg Tech conference in San Francisco. The 29-year-old founder of data-labeling startup Scale AI joined Meta after Chief Executive Officer Mark Zuckerberg overhauled his company’s AI efforts, including through a $14 billion investment in Scale that was widely viewed as a bid to recruit Wang. Since then, Wang has run Meta Superintelligence Labs, or MSL. The team, formed after he joined Meta, is beginning to show early results. In April, MSL debuted its first AI model since Zuckerberg embarked on the multibillion-dollar AI overhaul. Known as Muse Spark, it outperformed Meta’s previous AI. Wang conceded that Muse Spark is “not at the tier of the leading frontier models,” such as Anthropic PBC’s Claude and OpenAI’s ChatGPT, but that it was “better, frankly, than we expected internally.” He anticipates future models to be increasingly competitive. Health capabilities, Wang said, were one of Muse Spark’s greatest strengths. As he and his team continue to develop larger models, health will continue to be a focus — as could its integration into the company’s consumer-facing apps, including Instagram, Facebook and WhatsApp. As AI companies compete to build chatbots that resonate with consumers, health has emerged as one of the most closely watched business opportunities. Users are turning to ChatGPT, Claude and other top models in hopes of obtaining diagnoses, fitness tips and nutritional advice, and even mental health support — a phenomenon that has raised concerns about patient safety and privacy.Wang said that Muse Spark surfaced elevated biological-risk concerns during its development, though he didn’t describe them. While he added that Meta mitigated those risks before the model’s release, he said the finding helped inform a decision not to roll it out as “open source” — a process that makes the model’s building blocks broadly available to outside developers. ...moreElevate your knowledge and leadership skills at a cost cheaper than your daily tea.Subscribe Now