The OpenAI farce comes to an end, is Meta the "biggest winner" that has been overlooked?

The power struggle within OpenAI, the world's most famous tech startup, ended with CEO Ultraman taking back control within 100 hours. However, the behind-the-scenes battles between tech giants are far from over.

Some analysts point out that compared to Microsoft, which fought to avoid wasting billions of dollars in investments during the crisis, Meta, which was not directly involved in the power struggle, may be the silent winner. The reason is that the recent power struggle has made the market realize that exclusive models like OpenAI's GPT pose too much instability when tied too closely to a single company.

The Battle between Open Source and Closed Source

In the AI wave, there are two opposing factions. On one side, there is the "closed source" camp represented by OpenAI and Anthropic, where the technical details of their large models are not publicly disclosed, and the code is controlled by themselves. On the other side, there is the "open source" camp represented by Meta, where the code is open to the public and can be modified or even redeveloped within the scope of the license.

Supporters of the "open source" camp believe that open sourcing is beneficial for the rapid establishment and development of the ecosystem. It can also gather global strength to help AI models iterate quickly. Open sourcing is seen as opening up the wisdom of the crowd and decentralizing power to counter excessive centralization.

However, supporters of the "closed source" camp argue that open sourcing is harmful to creators. Competitors can replicate their work, and there may be potential dangers if advanced AI models are used for malicious purposes.

Ilya Sutskever, Chief Scientist of OpenAI, has explicitly stated:

"If you believe AGI is coming, open sourcing is a bad idea."

Mark Surman, President and Executive Director of the Mozilla Foundation, the developer of the Firefox browser and an internet open-source organization, told the media:

"The dramatic events of OpenAI/Microsoft highlight a major risk in recent artificial intelligence. The next wave of technology will be controlled by the small group of players who shaped the previous internet era... If GPT-X can be responsibly open-sourced, giving researchers and startups the opportunity to make this technology safer, more useful, and more trustworthy, then perhaps we have a chance to avoid this situation."

Just a few weeks ago, Mozilla, together with Yann LeCun, Chief AI Scientist of Meta, and other prominent figures, released an open letter calling for greater openness in the AI industry. The letter has garnered signatures from over 1,700 people.

The letter states:

"Increasing public access and scrutiny makes technology safer, not more dangerous. The idea that strict proprietary control over foundational AI models is the only way to protect us from large-scale societal harm is, at best, naive and, at worst, dangerous."

The Power Struggle may Impact OpenAI's Business

According to reports, over the weekend, hundreds of OpenAI customers have started reaching out to OpenAI's competitors, including Anthropic, Google, and Cohere, out of concern that their businesses may be affected if OpenAI were to disintegrate overnight.

Their concerns are not unfounded. The cloud computing industry is a typical example. Luis Ceze, a computer science professor at the University of Washington and CEO of OctoML, told the media:

"Putting all your eggs in one basket is very dangerous. We saw this in the early days of cloud computing, which led companies to turn to multiple vendors and hybrid environments."

Compared to OpenAI's GPT large models, Meta's open-source models have a relatively small gap in capabilities, making them a cost-effective and ideal channel for companies to achieve AI diversification. Companies that rely on proprietary models can quickly minimize business risks by deploying open-source models.

Furthermore, some analysts argue that the brief power struggle within OpenAI highlights the shortcomings of small enterprises in terms of governance. Some technology professionals who value stability may be more willing to join Meta's AI research laboratory.