How Meta’s Open Source AI is Giving US the Edge in AI Race

Countries are vying for AI dominance, with countries like the UK, India and China hot on America’s heels.

As AI evolves and becomes increasingly integrated into various sectors, from healthcare to finance, its potential impact on national security and economic competitiveness has come to the forefront of international discourse.

The US, recognising the strategic importance of AI, is taking steps to maintain its technological edge while promoting responsible development and deployment of these systems.

In this context, Meta, the technology company behind Facebook and Instagram, has announced that its open-source AI model, Llama, will be made available to US government agencies and contractors working on national security applications.

Meta's Llama Models: Explained

Meta's Llama is an open-source family of large language models, with the latest version Llama 3.1 featuring up to 405 billion parameters.

It supports eight languages and has a 128,000-token context window.

Meta’s Llama models are designed for various applications, including synthetic data generation, model distillation and multi-modal tasks.

Unlike closed models from OpenAI, Anthropic and Google, Llama can be freely downloaded and customised by developers.

Meta has partnered with major cloud providers like AWS, Google Cloud and Microsoft Azure to offer hosted versions of the model on their platforms.

Llama's open nature and competitive performance make it a significant player in the global AI landscape.

Open-source AI: A strategic asset for national security

Meta’s decision for Llama to be made available to US government agencies and contractors working on national security applications marks a significant shift from Meta's previous stance, which prohibited the use of its AI software for military purposes.

The company is partnering with several major technology and consulting firms to bring current and future Llama models to government agencies.

These partners include Accenture Federal Services, Amazon Web Services, Anduril, Booz Allen, Databricks, Deloitte, IBM, Leidos, Lockheed Martin, Microsoft, Oracle, Palantir, Scale AI and Snowflake.

Various applications of the Llama model are already being developed for national security purposes.

Oracle, for instance, is using Llama to synthesise aircraft maintenance documents, potentially reducing repair times for critical aircraft.

Scale AI is fine-tuning the model to support specific national security team missions, such as planning operations and identifying adversaries' vulnerabilities.

Nick Clegg, President of Global Affairs at Meta, summarises the potential impact of Llama for the US: “These kinds of responsible and ethical uses of open source AI models like Llama will not only support the prosperity and security of the US, they will also help establish US open source standards in the global race for AI leadership.”

Broader implications for US technological leadership

The deployment of open-source AI models like Llama in national security contexts is part of a broader strategy to maintain US technological leadership.

Mark Zuckerberg, Founder and CEO of Meta, says: “Today, several tech companies are developing leading closed models. But open source is quickly closing the gap.

Last year, Llama 2 was only comparable to an older generation of models behind the frontier. This year, Llama 3 is competitive with the most advanced models and leading in some areas.”

Meta argues that widespread adoption of American open-source AI models serves both economic and security interests, particularly in a world where national security is increasingly linked to economic output and innovation.

Meta suggests that such a standard is likely to emerge as open-source models become more capable and widely adopted, drawing parallels with the development of technologies like Linux and Android.

According to Meta, it is crucial for this standard to be rooted in democratic values and to support the strategic interests of the US and its allies.

The company states that open-source development can lead to faster innovation, reduced costs and better products, thanks to contributions from developers worldwide.

Challenges ahead

However, Meta's decision is likely to attract scrutiny.

The military utilisation of Silicon Valley technology has sparked controversy in recent years, with employees from companies such as Microsoft, Google and Amazon publicly opposing certain agreements with military contractors and defence agencies.

Additionally, Meta has faced criticism regarding its open-source AI strategy.

While OpenAI and Google contend that their AI technology is too potent and prone to misuse to be released publicly, Meta argues that AI can only be enhanced and rendered safer by allowing extensive public scrutiny of the code.

This move by Meta comes at a strategic moment, as AI companies contend with mounting regulatory challenges and privacy concerns.

It also unfolds amid the ongoing US-China trade war, marked by strict controls on AI technology.

For the future of Llama, Mark concludes: “Starting next year, we expect future Llama models to become the most advanced in the industry.

But even before that, Llama is already leading on openness, modifiability and cost efficiency.”

Make sure you check out the latest edition of Technology Magazine and also sign up to our global conference series - Tech & AI LIVE 2024

Technology Magazine is a BizClik brand