Meta Platforms Just Launched Revolutionary Artificial Intelligence (AI) — and It Could Be Dangerous | The Motley Fool

The race to create the best artificial intelligence (AI) models is definitely on, with current leader OpenAI making waves in late 2022, and promising competitors, Alphabet‘s (GOOG -0.28%) (GOOGL -0.17%) Gemini and AI startup Anthropic are also building high-performance models with hundreds of billions of parameters.

However, investors should not overlook Meta-platforms (META 2.71%)which takes the most unique and differentiated approach to creating generative AI models.

Will this approach give Meta a head start in the industry? Or do the risks outweigh the benefits?

When competitors close, Meta opens

Last week, Mark Zuckerberg and Meta celebrated the release of Meta’s Llama 3.1 model, which includes 405 billion parameters.

It’s a huge model. Until now, Meta had only released Llama models with 8 billion parameters and 70 billion parameters. But Llama 3.1 405B is Meta’s first salvo into “frontier” models, pushing the boundaries of the most advanced large-scale generative AI language models (LLMs) on the planet. According to Meta, the new model outperforms even the most advanced LLMs, such as ChatGPT 4-Omni and Anthropic’s Claude 3.5 Sonnet model, on many, but not all, parameters. Zuckerberg also claims that Llama 3.1 inference costs only 50% of the cost of running ChatGPT 4o.

Not only does Llama offer performance comparable to or better than many of the best models on the market today, it also has the distinction of being the only open source frontier model.

Benefits of Open Source

Open source software means that the licensee allows the software’s source code to be freely accessible and modified by third parties. External developers can then make changes to the software, allowing them to improve functionality, fix bugs, or improve security.

By giving away the software code for free, developers have the ability to improve the product more quickly than with a “closed” software system where only employees of a company have access to modify the code. Since external developers are turning to “free” open source software rather than expensive proprietary software, the open source model is best if you want to expand usage quickly. In addition to these general benefits of open source, Llama can run anywhere, so it doesn’t force developers to send their private data to a closed model or specific cloud.

This is why Mark Zuckerberg believes that Meta AI will become the most used AI assistant in the world by the end of this year, surpassing OpenAI.

In his blog post accompanying the license’s release, Mark Zuckerberg wrote that he saw the development of LLMs progressing in a similar way to the development of Linux and Unix operating systems in the 1990s. Although the closed Unix system took the lead early on, the advantages of Linux’s open source model eventually paved the way for its adoption as the standard operating system for cloud computing and mobile devices.

Zuckerberg also notes that Meta has benefited from open source tools in the past, such as the development of open source data center architecture and artificial intelligence software like Pytorch, which Meta originally developed. Since Meta’s primary business comes from advertising and not directly monetizing software or data center architecture, open source hasn’t hindered its ability to generate revenue and invest in research and development. At the same time, Meta has ended up saving billions of dollars by allowing these tools to be developed with the help of outside parties and then using them to run its main social media platforms.

The same goes for Llama AI. Meta is not a software vendor or a public cloud provider, so it doesn’t really rely on selling its model directly to generate revenue.

But Meta should take advantage of Llama… eventually

That’s not to say Meta doesn’t plan to generate revenue from Llama. But it will likely do so by taking the underlying model and building services on top of it. In its recent Q1 conference call with analysts, Zuckerberg hinted at building enterprise messaging customer service AI on WhatsApp, introducing ads into AI interactions in Meta AI, or perhaps charging for access to larger AI workloads with more computing power.

Other open source software companies, like Red Hat, have already monetized their AI models by selling consulting and customer support services. WordPress used a dual-licensing model to monetize its website builder software, offering a “deluxe” paid version to large enterprises. So there’s precedent for leveraging open source models.

Meta likely won’t see any direct revenue from Llama in the near term, but over the next few years the company should begin to deploy and benefit from ancillary services built on the underlying Llama model.

The inconvenients

But open source also carries risks: especially in the context of AI. These risks relate to privacy and security, elements for which Meta has been criticized in the past. Concerns about privacy and security are likely the reason why discussions with Apple (AAPL 0.22%) Plans to integrate Llama into Apple’s new Intelligence platform fell through earlier this year. Apple, which has perhaps the most closed business model in tech, still places a high value on privacy and security. So the disagreement with Meta and its exclusion from Apple Intelligence could be a red flag.

In particular, there may be reluctance to openly release the model source code and weights, as this could empower malicious actors who would like to modify the code for malicious purposes. This risks intervention by government authorities, who could limit the distribution of Llama to prevent it from falling into the hands of hostile governments, such as China, Russia, or Iran.

Zuckerberg responds that these countries are good at spying and will be able to access the best models anyway, even in a closed system. This essentially means that fewer “good guys” will have access to the best models (only a few large companies), which risks putting the United States at a disadvantage. Furthermore, Zuckerberg believes that stifling open innovation risks depriving the United States and its allies of the best AI.

But this argument seems unconvincing: Since these malicious actors will have access to the leading AI models anyway, should we just give them the code for free? It is unclear whether this argument could actually convince the US government to follow this line of thinking.

A major factor in the AI ​​race

Meta’s move to open source AI gives it the opportunity to become the world’s leading AI company and, as a result, perhaps one day the world’s most valuable company. But this approach also risks government oversight and regulation, threatening its ability to compete in the AI ​​war.

For Meta investors and AI tech investors in general, the rollout of Llama 3.1 and the potential government response to it in an election year are key things to watch in the coming months.