OpenAI, Google, Meta, Amazon and others pledge to watermark AI content for safety

AI companies, including OpenAI, Alphabet (parent company of Google), and Meta Platforms (formerly Facebook), have taken voluntary steps to enhance the safety of artificial intelligence technology. US President Joe Biden announced these commitments during a White House event aimed at addressing concerns regarding the potential misuse of AI and its impact on US democracy.

Biden acknowledged the importance of these commitments as a positive step but emphasised that there is still much work to be done collaboratively. He stressed the need to be vigilant about the threats posed by emerging technologies, particularly AI, to safeguard national security and democratic values.

The companies involved in these voluntary commitments also include Anthropic, Inflection, Amazon, and Microsoft, which is a partner of OpenAI. They have pledged to rigorously test AI systems before release, share information on risk reduction measures, and invest in cybersecurity to protect against potential attacks.

The move signifies a significant development in the Biden administration's efforts to regulate AI technology, which has experienced substantial investment and popularity among consumers in recent times. Microsoft, in response to Biden's leadership, expressed support for the collaborative effort to make AI safer, more secure, and beneficial for the public.

One of the main concerns addressed in the commitments is the rise of generative AI, which employs data to create new content, such as human-like prose produced by ChatGPT. Lawmakers worldwide have been increasingly contemplating ways to mitigate the risks posed by this rapidly emerging technology to national security and the economy.

It is worth noting that the United States has been lagging behind the European Union (EU) in terms of AI regulation. In June, EU lawmakers reached an agreement on draft rules that require AI systems like ChatGPT to disclose AI-generated content, distinguish deep-fake images from real ones, and implement safeguards against illegal content.

In response to calls for comprehensive legislation from US Senate Majority Chuck Schumer, Congress is currently considering a bill that would mandate political ads to disclose whether AI was used in creating their imagery or content.

To strengthen efforts in regulating AI, Biden has been actively working on developing an executive order and bipartisan legislation focusing on AI technology. He believes that the next few years will witness an unprecedented technological transformation, surpassing any changes seen in the past five decades.

Also Read Hollywood vs AI: Why famous actors including Oppenheimer, Barbie cast are on strike

As part of their commitments, the seven companies have vowed to develop a watermarking system that can be applied to all forms of content generated by AI, including text, images, audio, and videos. The watermark will be technically embedded in the content, enabling users to identify when AI technology has been used to create it.

This watermarking initiative aims to aid users in recognizing deep-fake images or audio, which might depict non-existent violence, facilitate scams, or manipulate images of politicians negatively. However, the specifics of how the watermark will be evident during the sharing of information remain unclear.

Additionally, the companies have committed to focusing on preserving user privacy as AI technology advances and ensuring that AI systems are free from bias, with measures in place to prevent discrimination against vulnerable groups. The commitments also extend to the development of AI solutions to address scientific challenges like medical research and climate change mitigation.

Also Read

Battle of the billionaires: Elon Musk vs Mark Zuckerberg cage match could make over $1 billion

Google appeals to Supreme Court to quash antitrust directives on Android in India