When Meta Chief Executive Mark Zuckerberg announced last year that his company would release an artificial intelligence system, Jeffrey Emanuel had reservations.
Mr Emanuel, a part-time hacker and full-time AI enthusiast, had tinkered with “closed” AI models, including OpenAI's, meaning the underlying code of the systems could not be accessed or modified. When Mr. Zuckerberg introduced Meta's AI system by inviting only a few academics, Mr. Emanuel worried that the technology would be limited to a small circle of people.
But in last summer's release of an updated AI system, Mr. Zuckerberg made the code “open source” so that anyone can freely copy, modify and reuse it.
Sold to Mr. Emanuel, founder of blockchain startup Pastel Network. He said he appreciates that Meta's AI system is powerful and easy to use. Most of all, he liked how Mr. Zuckerberg was supporting the hacker code of making the technology freely available — the exact opposite of what Google, OpenAI and Microsoft have done.
“We have this champion in Zuckerberg,” Mr. Emanuel, 42, said. “Thank God we have someone protecting open source ethics from these other big companies.”
Mr Zuckerberg has become the most high-profile tech executive to support and promote an open-source model of AI, putting the 40-year-old billionaire at the forefront of a controversial debate over whether the technology is potentially world-changing. There is too much. It's dangerous to be available to any coder who wants it.
Microsoft, OpenAI and Google have a more closed AI strategy to protect their tech, with an abundance of caution in what they say. But Mr. Zuckerberg stands strongly behind how technology should be open to all.
“This technology is so important, and the opportunities are so great, that we should open source it and make it as widely available as we responsibly can, so that everyone can use it,” he said in an Instagram video in January. can take advantage of.”
That stance has turned Mr. Zuckerberg into the man of the moment in many Silicon Valley developer communities, prompting talk of a “glow-up” and a “Zuckaissance” of sorts. Even as the chief executive continues to scrutinize misinformation and child safety issues on Meta's platforms, many engineers, coders, technologists and others have voiced their concerns about making AI available to the public. Accepted position.
Since Meta's first fully open-source AI model, called LLaMA 2, was released in July, the software has been downloaded more than 180 million times, the company said. A more powerful version of the model, LLaMA 3, released in April, reached the top of the download charts on Hugging Face, a community site for AI code, at record speed.
Developers have built thousands of their own custom AI programs on top of Meta's AI software to perform everything from helping clinicians read radiology scans to scoring digital chatbot assistants.
“I told Mark, I think open sourcing LLaMA is the most popular thing Facebook has done in the tech community — by far,” said Patrick Collison, chief executive of payments company Stripe, who recently I have joined the Meta Strategic Advisory Group. The aim is to help the company make strategic decisions about its AI technology. Meta owns Facebook, Instagram and other apps.
Mr. Zuckerberg's newfound popularity in tech circles is surprising given his rich history as a developer. Over two decades, Meta has sometimes pulled the rug out from undercoders. In 2013, for example, Mr. Zuckerberg bought Parse, a company that built developer tools, to attract coders to build apps for Facebook's platform. Three years later, he shut down the effort, angering the developers who had poured their time and energy into the project.
Mr. Zuckerberg and a spokeswoman for Meta declined to comment. (The New York Times filed a lawsuit against OpenAI and its partner Microsoft last year, claiming copyright infringement for news content about AI systems.)
Open source software has a long and storied history in Silicon Valley, with major tech battles revolving around open versus proprietary — or closed — systems.
In the early days of the Internet, Microsoft strove to provide the software that ran the Internet's infrastructure, only to eventually lose out to open-source software projects. More recently, Google open-sourced its Android mobile operating system to compete with Apple's closed iPhone operating system. Firefox, an Internet browser, WordPress, a blogging platform, and Blender, a popular set of animation software tools, were all built using open source technologies.
Mr. Zuckerberg, who founded Facebook in 2004, has long supported open-source technology. In 2011, Facebook launched the Open Compute Project, a nonprofit that freely shares the design of servers and equipment inside data centers. In 2016, Facebook also developed Pytorch, an open-source software library that has been widely used to build AI applications. The company is also sharing blueprints of computing chips it has developed.
“Mark is a great student of history,” said Daniel Eck, Spotify's chief executive, who counts Mr. Zuckerberg as a confidant. “Over time in the computing industry, he's seen that there are always closed and open paths to go. And he's always made the mistake of opening.
At Meta, his decision to open source his AI was controversial. In 2022 and 2023, the company's policy and legal teams backed a more conservative approach to releasing the software, fearing a backlash from Washington and EU regulators. But meta-technologists such as Yann Lacon and Joel Pino, who lead AI research, pushed for an open model, which they said would better serve the company in the long run.
The engineers won. Mr. Zuckerberg agreed that if the code was open, it could be improved and made safer faster, he said in a post on his Facebook page last year.
While open-sourcing LLaMA means giving away the computer code that Meta spent billions of dollars to build without an immediate return on investment, Mr. Zuckerberg calls it “good business.” As more developers use Meta's software and hardware tools, the more likely they are to invest in its technology ecosystem, which helps strengthen the company.
The technology has also helped Metta improve its internal AI systems, helping with ad targeting and more relevant content recommendations on Metta's apps.
“It's 100 percent tied to Zuckerberg's incentives and how it can benefit Meta,” said Noor Ahmed, an MIT Sloan researcher who studies AI, adding that “LLaMA is a win-win for everyone.”
Competitors are taking note. In February, Google open-sourced the code for two AI models, Gemma 2B and Gemma 7B, a sign that it was feeling the heat from Mr. Zuckerberg's open-source approach. Google did not respond to requests for comment. Other companies including Microsoft, Mistral, Snowflake and Databricks have also started offering open source models this year.
For some coders, Mr. Zuckerberg's AI approach hasn't erased all the baggage of the past. Sam McLeod, 35, a software developer in Melbourne, Australia, deleted his Facebook accounts years ago after growing unease with the company's track record on user privacy and other factors.
But more recently, he said, he recognized that Mr. Zuckerberg had released “cutting-edge” open-source software models with “permissive licensing terms,” which could not be said for other big tech companies. .
Matt Schumer, 24, a developer in New York, said he used closed AI models from Mistral and OpenAI to power digital assistants for his startup HyperRight. But when Meta released its updated open-source AI model last month, Mr. Schumer began relying heavily on it instead. Any reservations about Mr. Zuckerberg are in the past.
“Developers are starting to see a lot of the problems that they and Facebook had in the past,” Mr. Schumer said. “Right now, what he's doing is really good for the open source community.” “