Microsoft President Brad Smith urges the US Congress to pass a new "deepfake fraud statute"

With the rise of generative AI has come a new wave of digital "deepfake" content by malicious actors who are trying to get false information about people and topics on the internet. That trend has become especially worrisome because of the 2024 US Presidential Election.

On that particular front, the Microsoft Threat Analysis Center (MTAC) has been monitoring how state-supported threat actors from Russia, Iran, and China have been creating false websites, social media posts, and videos to try to influence the 2024 election. Indeed, earlier this week, the company published a new report claiming that Russia-based actors were making videos that were designed to push out false information about the Democratic US Presidential nominee Kamala Harris and its Vice-Presidential nominee Tim Walz.

Today, Microsoft posted the written testimony of company president Brad Smith on its blog, as he also appeared before the US Senate Intelligence Committee in Washington DC. The hearing was centered on foreign threats in US elections and the responsibility of US tech companies like Microsoft to deal with these threats.

In his written testimony, Smith repeated the many actions from state-sponsored groups to use AI and deep fakes to influence the 2024 elections. He also urged the committee and the US Congress to take action themselves.

In his testimony, Smith wrote:

First, Congress should enact a new federal “deepfake fraud statute.” We need to give law enforcement officials, including state attorneys general, a standalone legal framework to prosecute AI-generated fraud and scams as they proliferate in speed and complexity.

Smith also said that Congress should also pass a requirement that any AI system provider must label any AI-created content with "state-of-the-art provenance tooling". Finally, Smith said the Congress should pass the Protect Elections from Deceptive AI Act, which has already been proposed by a bipartisan group of US lawmakers. Among other things, the bill would prevent the use of AI in generating political content for federal lawmakers, with exceptions made for satirical content or for use in newsrooms.