Meta’s Helping the US Government Integrate Its Llama AI Models

As part of Meta’s Q3 earnings report last week, CEO Mark Zuckerberg made a brief note that raised many eyebrows in response.
In the earnings call, Zuckerberg noted that:
“This quarter, we released Llama 3.2, including the leading small models that run on-device and open source multi-modal models. We’re working with enterprises to make it easier to use, and now we’re also working with the public sector to adopt Llama across the US government.”
Meta’s working with the government on the adoption of AI in government applications? What could that mean?
Well, today, Meta has provided more context, with its head of Government Affairs Nick Clegg explaining that Meta is indeed working with U.S. government agencies on potential applications for its Llama AI models, while also partnering with some big-name corporations on the same.
As explained by Clegg:
“Meta’s open source Llama models are increasingly being used by a broad community of researchers, entrepreneurs, developers and government bodies. We are pleased to confirm that we are also making Llama available to U.S. government agencies, including those that are working on defense and national security applications, and private sector partners supporting their work. We’re partnering with companies including Accenture, Amazon Web Services, Anduril, Booz Allen, Databricks, Deloitte, IBM, Leidos, Lockheed Martin, Microsoft, Oracle, Palantir, Scale AI and Snowflake to bring Llama to government agencies.”
So the generative AI systems that regularly misinterpret facts from the web, and convert them into convoluted, unreal narratives, are now going to be powering elements for the Defense Department. Got it, sounds good.
Nah, it should be fine. I mean, it’s not like they’re going to use it for mission planning or anything major like that.
“Scale AI is fine-tuning Llama to support specific national security team missions, such as planning operations and identifying adversaries’ vulnerabilities.”
Oh.
I mean, that seems like a concern.
Overall, however, Clegg says that Meta’s AI models are being tested in applications where they can simplify data analysis and interpretation.
“[Large language models] can help to streamline complicated logistics and planning, track terrorist financing or strengthen our cyber defenses. For decades, open source systems have been critical to helping the United States build the most technologically advanced military in the world and, in partnership with its allies, develop global standards for new technology.”
The adoption of the latest AI models, then, is in line with keeping the U.S. government a step ahead of potential enemies, and using the best tools available to improve its operations.
So it makes sense. Nothing to be concerned about.
I mean, it’s not like they’re going to develop robots and give over control of them to AI systems and those systems are then going to take over and determine that humans are the real enemy and eliminate us all.
“In July, a former chairman of the Joint Chiefs of Staff said that robots and other smart machinery will comprise up to one-third of the U.S. military in the next 10-15 years.”
…
Okay then.
Well, I mean, we all have enough to be stressed about this week, so maybe don’t think too hard about Meta giving over AI models to the U.S. Defense Department, which is also developing autonomous robots equipped for the battlefield.