Meta’s open-source Llama models are gaining traction among a diverse group of users, including researchers, entrepreneurs, developers, and government organizations. We’re excited to announce that we’re also extending access to Llama for U.S. government agencies, especially those involved in defense and national security efforts, as well as private partners who are assisting them.
We’re teaming up with notable companies like Accenture Federal Services, Amazon Web Services, Anduril, Booz Allen, Databricks, Deloitte, IBM, Leidos, Lockheed Martin, Microsoft, Oracle, Palantir, Scale AI, and Snowflake to make Llama accessible to these agencies.
For instance, Oracle is leveraging Llama to create aircraft maintenance documents that help technicians diagnose issues more swiftly and accurately—this means faster repairs and getting essential aircraft back in the air sooner.
Scale AI is customizing Llama for specific missions related to national security like operational planning and pinpointing vulnerabilities of adversaries. Lockheed Martin has integrated Llama into its AI Factory to enhance various tasks such as code generation and data analysis while streamlining business processes.
Additionally, Amazon Web Services and Microsoft Azure are using Llama in their secure cloud environments to assist government operations involving sensitive information. IBM’s watsonx solution is also integrating Llama for use by national security agencies within their own data centers or cloud setups.
These responsible applications of open-source AI models like Llama not only bolster the safety and prosperity of the United States but also help set U.S. open-source standards in the global race for leadership in AI technology. As an American company that thrives on the entrepreneurial spirit and democratic values our country stands for, Meta is committed to contributing towards America’s safety and economic growth—alongside its closest allies.
Large language models have a knack for processing massive amounts of data quickly while generating valuable insights—they can streamline complex logistics plans or track down terrorist financing or even enhance our cyber defenses. For many years now, open-source systems have played a vital role in helping the U.S. establish itself as a leader with one of the most advanced militaries globally; they’ve also supported our allies in creating global tech standards.
These systems have been key in accelerating defense research efforts while improving communication across various platforms by identifying security weaknesses along the way.
In today’s world, where national security is closely tied to economic performance, innovation, and job creation, the widespread use of American open source AI models benefits both our economy and security. Other countries, especially China and various U.S. competitors, are aware of this too and are in a race to create their own open source AI models, pouring in significant resources to get ahead of us. It’s crucial for both the U.S. and other democratic nations that American open source models thrive over those from China and elsewhere.
As these open source models become more advanced and gain traction globally, we’re likely to see a universal standard for AI emerge—similar to what happened with technologies like Linux and Android. This will unfold regardless of whether the U.S. takes part or not. Such a standard will underpin AI development worldwide, becoming integral to technology, infrastructure, manufacturing, global finance, and e-commerce.
It’s vital that this global standard is built on principles of openness, transparency, and accountability—this is where American leadership comes into play along with its commitment to international law for maintaining peace and security globally. Countries using AI for national security must ensure they deploy it ethically and responsibly while adhering to international laws—principles that the U.S. and many allies have embraced in the Political Declaration on Responsible Military Use of Artificial Intelligence.
The aim should be to create a positive feedback loop: helping the U.S. maintain its technological edge while spreading access to AI around the globe in a way that promotes responsible innovations aligned with the strategic interests of America and its allies.
Open source can play a key role here by fostering faster innovation at lower costs while improving product quality through contributions from thousands of developers worldwide. Both public sector entities in the U.S. as well as governments everywhere stand to gain significantly from open source AI access—it drives discoveries that enhance efficiency and help government employees deliver better public services.
For instance, Deloitte is introducing Llama-based solutions tailored for U.S. government agencies and nonprofits so they can better serve their communities across areas like education, energy initiatives, small business support, skilled trades training—and beyond! Collaborations between public entities and private organizations like this can really elevate operations while making a meaningful difference for those they serve.
Meta is teaming up with some influential figures in the tech world to champion AI systems that are not only safe and reliable but also focus on pressing social challenges like ensuring access to clean water and dependable electricity, while also lending a hand to small businesses.
A great example of this is Meta’s collaboration with UNESCO to develop a translation tool powered by our No Language Left Behind (NLLB) AI model. This tool aims to provide top-notch translations in 200 languages, including many that often get overlooked.
It’s truly exciting to see the global enthusiasm for the incredible potential of this accessible technology! By working closely with others, we’re striving to bolster America’s role in tech and make sure that the next generation of digital infrastructure around the globe is built on democratic values and protections.