Facebook owner Meta has released a new and free-of-charge version of its artificial intelligence [AI] model, making a play against ChatGPT-maker OpenAI and Google.
OpenAI and Google have developed impressive large language models that serve as the foundations of the ChatGPT and Bard chatbots, which have drawn excitement with their capabilities to mimic human creativity and expertise.
Meta, meanwhile, has avoided releasing generative AI products straight to the consumer and instead developed Llama, a language model specifically developed for researchers so that they could perfect it.
Crucially, Llama is open-source, meaning that its inner workings are available to all to be tinkered with and modified, unlike the headline-grabbing AIs developed by OpenAI and Google.
Those models, including OpenAI's world-leading GPT-4, are closed and proprietary, with the clients that use them denied access to their programming code or detailed answers as to how their data is handled.
"Open-source drives innovation because it enables many more developers to build with new technology," Meta CEO Mark Zuckerberg said in a Facebook post on Tuesday.
"It also improves safety and security because when software is open, more people can scrutinise it to identify and fix potential issues," he added.
The stress on safety also underlines a divergence from OpenAI's models, which have caused alarm by generating false information or by going off the rails in chatbot interactions.
The new, more powerful version of Meta's model, called Llama 2, would be available to any business for download or through Microsoft's Azure cloud service in a special partnership with the Windows maker.
The Microsoft tie-in comes on top of that company's major partnership with OpenAI, signalling Microsoft is attempting to diversify its AI offerings with products that put businesses in more control of their data and software.
Microsoft, which has been the most aggressive big tech player to enter the AI market, saw its share price skyrocket on Tuesday when it said it would be charging $30 per user, per month for an AI-enhanced version of Microsoft 365, its office platform.
This would be a significant price hike for its business customers and could potentially lead to a vast increase in revenue for Microsoft if AI is seen as a necessary cost going forward.
'We have no moat'
For Meta, a flourishing open-source ecosystem of AI tech built using its models could stymie rivals' plans to earn revenue off their proprietary technology, the value of which would evaporate if developers could use equally powerful open-source systems for free.
A leaked internal Google memo titled "We have no moat, and neither does OpenAI" lit up the tech world in May after it forecasted such a scenario.
Meta is also betting that it will benefit from the advancements, bug fixes and products that may grow out of its model, becoming the go-to default for AI innovation, as it has over the past several years with its widely-adopted open source AI framework PyTorch.
As a social media company, Zuckerberg told investors in April, Meta has more to gain by effectively crowd-sourcing ways to reduce infrastructure costs and maximise the creation of new consumer-facing tools that might draw people to its ad-supported services than it does by charging for access to its models.
"Unlike some of the other companies in the space, we're not selling a cloud computing service where we try to keep the different software infrastructure that we're building proprietary," Zuckerberg said.
"For us, it's way better if the industry standardises on the basic tools that we're using, and therefore we can benefit from the improvements that others make."