The AI Ticking Time Bomb: How can we safeguard Democracy from Disinformation at Scale?

Are we facing a pivotal moment in human history without really noticing it? The unprecedented arms race of the tech giants like Microsoft, Google, and Facebook to dominate the "Artifical Age" has also accelerated the debate around the potential downside of possibly the most remarkable invention so far in the 21st century.

Just watch this interesting and slightly disturbing exclusive interview on abc News with Open AI Co-Founder Sam Altman. Even he admits he can't predict where the recent development is headed and calls for political oversight.

The Future of Life Institute (FLI) published an open letter some days ago (including signatories like Elon Musk, Steve Wozniak and Yuval Noah Harari) calling for an immediate moratorium on giant AI experiments, highlighting the potential risks and negative consequences that could arise from unchecked AI development: https://futureoflife.org/open-letter/pause-giant-ai-experiments/ .

In response to this letter, it is crucial for politicians in democracies worldwide to act swiftly to address these threats, particularly disinformation at scale, to safeguard our societies, economies, and, lastly democratic values. The letter emphasizes the potential risks of AI systems such as the GPT-4 architecture, including their potential to create disinformation, amplify biases, and undermine trust in democratic institutions.

What if we can't tell really anymore what is real and what isn't? Just watch this Demo presentation of MetaHuman and imagine where we will be headed within the next months.

Disinformation-at-scale: Wet dreams for Autocrats and nightmares for free and open democratic societies

One of the most pressing concerns arising from advanced AI systems is their ability to generate disinformation at an unprecedented scale. Through the creation of highly convincing fake news, deepfakes, and other misleading content, AI-powered disinformation can erode public trust, stoke social divisions, and interfere with the democratic process.

Thinking Cambridge Analytica was bad? Imagine a customized built AI that can use an entire army of artificially created bots and deepfakes to flood the Internet and public discourse within minutes. AI algorithms can analyze vast amounts of data on individual users to create tailored disinformation campaigns that exploit people's beliefs, fears, and biases. By targeting specific groups or individuals with misleading information, these campaigns can manipulate voter behavior and swing election outcomes.

Forget the stupid click-and-retweet troll-bots of the past. AI-powered bots can be deployed to spread disinformation rapidly and widely across social media platforms, amplifying the reach of false narratives. These bots can also create the illusion of popular support for certain viewpoints or candidates, potentially swaying undecided voters.

So what do now?

To counter these threats immediately, politicians in our western democracies must act swiftly to develop and implement regulations that govern the development and safe use of AI technologies. These regulations should focus on the following areas:

1. Transparency: Requiring AI developers from the most relevant studios and tech companies to disclose the workings of their systems to regulators, in order to ensure that ethical guidelines and safety measures are being followed.
2. Building Oversight institutions: Creating legal frameworks to hold developers and operators of AI systems accountable for the impacts of their technologies on society. Just as the banking industry can pose systemic risks to the economy due to financial contagion (as we just saw with Silicon Valley Bank or the Credit Suisse disaster), AI development can also present systemic risks to society. Public oversight institutions can help identify, monitor, and mitigate these risks, ensuring that AI technologies do not lead to unintended consequences or exacerbate existing societal issues. In most countries they still have to be designed and built.
3. Collaboration: Fostering international cooperation among governments, AI developers, and civil society organizations to establish global norms and standards for AI development and use. The UN needs to act fast to adapt to the current speed of innovation to follow suit.
4. Education and Public Awareness: Investing in public education and awareness campaigns to inform citizens about the potential risks and benefits of AI, as well as the importance of discerning reliable information sources. What if an image, an audio file or a video cannot be taken as real? Where can I check its authenticity?
5. Support for Independent Research: Allocating resources to fund independent research or fact-checking tools on AI safety, ethics, and governance, to inform evidence-based policymaking.

The open letter from the Future of Life Institute should really serve as a stark reminder of the potential risks posed by unchecked AI development. In order to protect our societies from the threat of disinformation at scale and other negative consequences, it is imperative that we call our politicians to act swiftly to develop and implement comprehensive regulations governing AI technologies. By doing so, they can help ensure that AI is harnessed for the greater good, while minimizing its potential for harm. As Harari coined it in his recent article for the New York Times: "The first step is to buy time to upgrade our 19th-century institutions for an A.I. world and to learn to master A.I. before it masters us."

Further reading recommendation: