While comprehensive AI legislation is still in the works and might take years to form, the rapid growth of AI has raised concerns.
This also includes the White House, which is concerned about the potential for irresponsible development.
YOU MAY ALSO LIKE: Twitter: Anticipated Feature To Assist Hiring Organizations
To address this, the Biden administration has gathered “voluntary commitments” from seven major AI developers to pursue shared safety and transparency goals.
These commitments come ahead of a planned Executive Order on AI.
The companies involved in this non-binding agreement include OpenAI, Anthropic, Google, Inflection, Microsoft, Meta, and Amazon.
Representatives from these firms are meeting with President Biden to discuss their commitments.
It’s essential to note that these are purely voluntary practices, and there is no proposed rule or enforcement to hold companies accountable.
However, the commitments will likely be part of the public record. The companies have agreed to the following:
1. Conduct internal and external security tests on AI systems before release, including inviting external experts for adversarial “red teaming.”
2. Share information with the government, academia, and civil society on AI risks and mitigation techniques, such as preventing “jailbreaking” of AI systems.
3. Invest in cybersecurity and safeguards against insider threats to protect private model data, which is crucial not only for intellectual property protection but also to prevent malicious exploitation.
4. Encourage third-party discovery and reporting of vulnerabilities, possibly through bug bounty programs or domain expert analysis.
5. Develop robust watermarking or similar methods to mark AI-generated content.
6. Report the capabilities, limitations, and appropriate and inappropriate uses of AI systems, though obtaining straightforward answers on this matter may be challenging.
7. Prioritize research on societal risks, including systematic bias and privacy concerns.
8. Use AI to address significant societal challenges such as cancer prevention and climate change, while also keeping an eye on the carbon footprint of AI models.
In attendance were prominent figures in the involved organizations including
Brad Smith – President at Microsoft
Kent Walker – President at Google
Dario Amodei – CEO at Anthropic
Mustafa Suleyman – CEO at Inflection AI
Nick Clegg – President, Meta
Greg Brockman – President, OpenAI
Adam Selipsky – CEO at Amazon Web Services
While these commitments are voluntary, the possibility of an Executive Order in development may incentivize compliance.
The threat of such an order could lead to additional directives, such as the FTC scrutinizing AI products that claim robust security but fail to allow external testing.
The White House is keen on staying ahead of the next wave of transformative technology, having learned from being caught off guard by the disruptive capabilities of social media.
President Biden and Vice President Harris have engaged with industry leaders for advice on a national AI strategy.
Moreover, significant funding has been allocated to establish new AI research centers and programs.
YOU MAY ALSO LIKE: OpenAI Releases New Feature For ChatGPT Users
However, it’s worth mentioning that the national science and research apparatus is already ahead of the game.
This is demonstrated by the DOE and National Labs’ highly comprehensive research challenges and opportunities report on AI, although the information may be slightly outdated.
