State legislatures are taking the lead in addressing the growing threat of deepfakes in elections, as Congress has been slow to act. Deepfakes, which are manipulated media content created using AI, have become increasingly convincing and easier to create in recent years. This week, South Carolina lawmakers introduced legislation that would ban the distribution of deepfakes of candidates within 90 days of an election, joining several other states that have passed similar bills. However, there are concerns about enforcing these laws and the impact of rapidly advancing AI technology.
While social media platforms have measures in place to combat disinformation, such as Facebook’s disclosure requirement for political advertisers, these measures do not apply to posts shared by everyday users. In Congress, proposals to regulate AI-created deepfakes have stalled, although the U.S. House recently unveiled the No AI Fraud act to protect artists and actors from AI deepfakes and voice clones.
To detect deepfakes, AI detection tools have been developed, but they have proven mostly unreliable. However, there is hope that advancements in AI technology can be used to combat deepfakes. McAfee has announced a new technology that it claims is 90% effective at detecting maliciously altered audio in videos, while Fox and Polygon Labs have unveiled a blockchain protocol for media companies to watermark their content as authentic. Despite these efforts, the problem of deepfake detection may never be completely solved and could become a cat-and-mouse game between technology advancements and bad actors.
The threat of deepfakes extends beyond elections, with AI deepfake technologies being used for various malicious purposes, including scams and distributing harmful content. As the 2024 elections approach, the stakes are high, as not only Americans but nearly half of the global population will be heading to the polls.
In other news, OpenAI has launched the GPT Store and ChatGPT Team, allowing users to find and share custom versions of ChatGPT for niche uses. Microsoft has used AI to rapidly identify an alternative to lithium in batteries, potentially reducing the amount of lithium needed by 70%. Volkswagen plans to integrate OpenAI’s chatbot, ChatGPT, into its cars to enhance voice control capabilities. Pennsylvania has become the first state to adopt ChatGPT Enterprise for state employees’ work tasks.
Additionally, there is growing interest in Microsoft’s relationship with OpenAI from European regulators. AI’s impact on the workforce is also a topic of discussion, with concerns that black workers could be left behind in AI wealth creation, exacerbating inequalities. A profile of Anna Makanju, OpenAI’s vice president of global affairs, sheds light on her strategic role in positioning OpenAI’s CEO, Sam Altman, as an ambassador for the AI industry.
Overall, the battle against deepfakes continues, with state legislatures taking action while Congress lags behind. AI technology shows promise in detecting deepfakes, but challenges remain. Meanwhile, OpenAI’s GPT Store and ChatGPT are expanding their capabilities, and Microsoft is exploring alternatives to lithium in batteries. The integration of AI in various industries, from automotive to government, is on the rise.
