Artificial intelligence (AI) is now pervasive in our lives. In fact, it’s been going on mostly under the hood for a while now.

For the past decade, we’ve used AI every time we search the internet or watch a movie on Netflix. It’s there when we use navigation apps to get from point A to point B, or when we use camera filters to give us smooth skin and bunny ears.

But it’s the emergence of “generative” AI in the last few years that has made it clear just how radically world-changing this technology will be.

Very simply, generative AI is AI that can create new things, such as text, images, videos, or even computer code, based on the examples shown so far.

Some fun examples have emerged – who will forget deepfake tom cruise again pope in down jacket?

A wave of technology-driven innovation culminated last year with the release of ChatGPT and image creation tools such as Stable Diffusion and Midjourney. These put the power of generative AI at your fingertips, even for those with little or no technical expertise.

This posed a rather serious dilemma. The world is already affected by misinformation and fake news, and the use of technology to spread slander and malicious falsehoods is becoming more and more common. Now that everyone has this technology (and it will no doubt become more sophisticated over time), how do we know again if what we see or hear is real? Is it possible?

The Age of Misinformation

“A lie can go halfway around the world while the truth still has its shoes on.” The thing is, this saying has never been more true than it is today.

From the internet to social media to “deepfakes” to generative AI, with the continuing wave of technological advancements, it’s becoming increasingly difficult to know for sure if what we see with our own eyes is real. It’s getting harder.

The US presidential election, the global COVID-19 pandemic, the UK’s referendum to leave the EU, and Russia’s invasion of Ukraine are all events of global significance. And they were all marked by concerted attempts to influence their development through the spread of targeted disinformation campaigns.

Deepfakes are certainly one of the most concerning products in the generative AI revolution. It’s now very easy to make it look like someone is saying or doing something, even if it would never happen in real life. . This can range from attempts to ridicule or embarrass politicians, to creating non-consensual pornography featuring celebrities and “revenge pornography” aimed at individuals. There are also instances where fake voices have been exploited by scammers. trick people out of cash By making you believe that your loved one is in trouble and needs help.

Deepfakes and other “creative” uses of generative AI can be misleading and distort reality without malicious intent. Hundreds of AI-generated songs that have never been sung by a human voice have flooded social media since the tool was put into the hands of the public. You will hear a “”.Beatles last recordA song featuring John Lennon (officially endorsed by the rest of the members), or Kurt Cobain’s cover of Blur’s Song 2 (less official).

Given the potential consequences, it is clear that a full understanding of the ethical implications of this whole-of-society transformation is of critical importance. How can we meet these challenges without impeding the undeniable opportunities for positive change and progress brought about by AI?

trust and regulation

The inability to distinguish between reality and AI-generated fantasies and lies can have disastrous consequences, both for politically important events and for our personal relationships.

Trust is essential in many areas of life. We must trust our elected leaders, we must trust our friends and loved ones, and the developers of AI tools that use our data to make decisions that affect our lives. must be trusted.

Perhaps most importantly, I need to be able to trust what I see and hear in order to be able to judge who else I can trust.

Regulators play a key role in determining whether AI can be trusted. In order to maintain public trust, it is hoped that laws aimed at preventing technology from being used for deceptive or misleading purposes will be enacted. Examples include laws such as: Enforced In January 2023, China will ban the use of deepfakes and AI techniques that could disrupt its economy and national security. The law also prohibits deepfake content featuring real people without their consent and imposes obligations on creators. synthetic content Just to make it clear that it’s not real.

However, there is always the danger that such methods may have the effect of stifling some of the innovation potential that AI will bring. The Chinese way of tackling this challenge through enforcement and regulation is therefore one possible solution to the problem, although other jurisdictions may prefer to adopt a more organic approach.

Other safety measures and solutions

Regulation is likely to play an important role in societal responses, but Rise of the Unrealother methods are probably just as important.

Perhaps embracing all this is the concept of digital literacy. We develop (and impart to others) the skills to critically evaluate the digital content we see, fostering a more resilient society against misinformation. Was it really possible for Kurt Cobain to cover Song 2?Could Joe Biden Really Be That Way? sang baby shark After introducing it as “our national anthem”?

Human fact-checking efforts can provide an additional line of defense. This includes educated and trained professionals who are adept at using scientific methods to ascertain facts from fiction and (as far as possible) operating without political or other types of prejudices. The house team gets involved. They will become increasingly important as misinformation technology evolves and becomes more sophisticated.

And, of course, technology itself can play a big role. Software-based solutions for detecting when content has been created or altered by AI are already available, with text-based generative AI of the type mass-produced by ChatGPT and videos used to create deepfakes, It covers both images and sounds.

Some form of regulation is certain, but these organic methods of combating the rise of disinformation can also be effective, thwarting genuine attempts to use generative AI for positive purposes. There’s also the added bonus that the odds aren’t that great.

Benefits and dangers of AI

Even if we, as individuals and as a society, are determined to tackle the potential for AI-powered misinformation, it is clear that we will face difficult challenges along the way. Employing a comprehensive strategy that includes critical thinking, fact-checking, and technical solutions will most likely mitigate danger while paving the way for good work.

Undoubtedly, overcoming the challenges discussed here is a balancing act, but we all have every incentive to do it. No matter how it unfolds, the next five years will be critical as society adapts to the presence of this world-changing technology and embraces its impact. This means that users, creators, legislators, beneficiaries, and anyone involved in AI in any way cannot afford to ignore the big question at the heart of this issue.