AI innovation presents global risks that outweigh nuclear war.

Many experts believe that alongside global opportunities, AI poses global risks, which will be greater than, say, the risks of nuclear technology - which in any case have historically been underestimated. Furthermore, scientific risk analysis suggests that high potential damages should be taken very seriously even if the probability of their occurrence were low.

While the risks from domain-specific AIs appear limited in the near future, there are long-term developments to take into consideration: in the not-so-distant future, artificial intelligence could in principle pose an existential threat, similar in scope to the pandemic risks associated with biotechnology [4, 6, 17].

<<<FIGURE OMITTED>>>

While the entertainment industry does offer significant opportunities for better education through personalized AI teaching and the gamification of learning material [66], it also increases the risk that a growing proportion of young people will have trouble completing their education due to a pathological addiction to video games and/or the internet [3].

On terrorism, crime, and other sources of risk:

The dangerous applications for AI would be criminals or large terrorist organizations using it to disrupt large processes or simply do pure harm. Terrorists could cause harm via digital warfare, or it could be a combination of robotics, drones, with AI and other things as well that could be really dangerous.

And, of course, other risks come from things like job losses. If we have massive numbers of people losing jobs and don’t find a solution, it will be extremely dangerous. Things like lethal autonomous weapons systems should be properly governed - otherwise there’s massive potential of misuse.

But this is the duality of this technology. Certainly, my conviction is that AI is not a weapon; AI is a tool. It is a powerful tool, and this powerful tool could be used for good or bad things. Our mission is to make sure that this is used for the good things, the most benefits are extracted from it, and most risks are understood and mitigated.

Some of the main risks associated with AI include:

Algorithmic bias: Machine-learning algorithms identify patterns in data and codify them in predictions, rules and decisions. If those patterns reflect some existing bias, the algorithms are likely to amplify that bias and may produce outcomes that reinforce existing patterns of discrimination.

Overestimating the capabilities of AI: Since AI systems do not understand the tasks they perform, and rely on their training data, they are far from infallible. The reliability of their outcomes can be jeopardized if the input data is biased, incomplete or of poor quality.

Programmatic errors: Where errors exist, algorithms may not perform as expected and may deliver misleading results that have serious consequences.

Risk of cyber attacks: Hackers who want to steal personal data or confidential information about a company are increasingly likely to target AI systems.

Legal risks and liabilities: At present, there is little legislation governing AI, but that is set to change. Systems that analyze large volumes of consumer data may not comply with existing and imminent data privacy regulations, especially the EU’s General Data Protection Regulation.

Reputational risks: AI systems handle large amounts of sensitive data and make critical decisions about individuals in a range of areas including credit, education, employment and health care. So any system that is biased, error-prone, hacked or used for unethical purposes poses significant reputational risks to the organization that owns it.

In the example of autonomous cars, the blind use of deep neural nets (DNNs) coupled directly to vehicle action control systems would be very risky: it would be equivalent to asking a taxi driver who has lost more than 80% of his brain as a result of an accident (and only keeping this occipito-temporary lane) to drive a car. It is simply not possible to ask these systems to do more than what they were originally designed for, at the risk of producing dramatic accidents.

Artificial intelligence brings with it not only enormous potential, but also a number of challenges, fears and risks.

There are five types of major risks of AI: bugs, cybersecurity, the "Sorcerer's Apprentice" phenomenon, shared autonomy, and possible negative socio-economic effects. Bugs refer to programming errors in the AI software. Just as there are dangers from cyberattacks, this risk is not different from current IT systems. "Sorcerer's Apprentice" risks refer to AI systems that perform incorrect actions when unclear instructions are present from the user.

Not a few experts are of the opinion that, in addition to global opportunities, the AI also poses global risks, which have been historically underestimated for a long time, such as those of nuclear technology. In addition, a knowledge-based risk analysis suggests that high potential damage levels should be taken very seriously even if the probabilities of occurrence were low.