Information Systems Integration – Tony Messina

The Ethics of Overpaying AI Talent Salaries

The lucrative field of artificial intelligence has countless applications in today’s world, from handwriting, speech, and facial recognition to huge projects such as Uber’s self-driving cars and Google’s Alpha Go, which is a program that can learn without the help of humans. However, according to a lab in Montreal, there are less than 10,000 people in the entire world who possess the knowledge and experience necessary to work on serious artificial intelligence projects– making each of these people worth enormous sums of money. Nowadays, tech giants such as Uber and Google are paying their AI specialists hundreds of thousands up to millions of dollars in stock and salary due to the gap in supply and demand, which raises the question of how smaller startups and business will compete. Even with the promise of owning a part of stock in a growing company, small startups cannot compete with the enormous salaries AI engineers and even professors (who are being taken out of the classroom to work in industry) are being offered. This effectively locks anyone else out of the industry and creates an oligopoly within the AI industry, so the question is — is this a fair situation to smaller companies? Should we make more of an effort to accelerate AI teaching throughout the world to scout talent? Or should we take the initiative to put salary caps on AI specialists like the NFL?


3 Responses to The Ethics of Overpaying AI Talent Salaries

  • It’s really a good and interesting article! I think start-up companies also have their own privileges. Of course, giant technology companies will pay a big amount of salaries, but start-ups could provide company shares and pay packages to the AI talents. If the start-up is going to be successful in the future, the AI talent will earn more than the salaries that giant technology companies could pay. In conclusion, I think “fair” does not really matters under this circumstance.

    I think it is hard to accelerate AI teaching since most of the top talents for AI must have a Ph.D. degree. I remember when I reading the article that the giant companies are attracting AI Talents who teach in University to become a full-time contractor for them, which will have a bad influence on the AI teaching. If the AI study could attract more talent, it would be the best! In addition, I don’t think we could set salary caps for AI talents now because they are really scarce.


  • I think current high salary for AI experts is the result of the market, in other words, the AI experts deserve the high salary. It is true that the traditional small companies can’t afford that huge amount expense, but there are many startups are actually found by those AI specialists. As people view the artificial intelligence technology as next disruptive innovation, some visionary AI specialists are confident enough to give up the high salary in the big company and start their own business for long-term profit maximization. Also, maybe the oligopoly within the AI industry in short-term is not a bad thing. Current AI is still far from maturity, the oligarch can concentrate and integrate the resources for better AI development, and the society as a whole can benefit from the AI applications that produced by the oligarch. So I think right now we shouldn’t regulate the AI industry in terms of the monopoly or oligopoly.

  • By taking the experts out of the educational system, it may limit the supply of qualified graduates in the future. However, the high salaries and attractiveness of the AI positions and the opportunity for innovation at large firms will only attract more talent to the specialty. The incentive is high and bringing in more people to work on AI projects will contribute to the overall knowledgebase and maturity of the technology. As Run said, the high salaries are market driven – the low supply and relatively high demand drives up salaries. The market SHOULD determine how much the experts get compensated based on how much it values the expertise, which is clearly a lot. Capping that will decrease the incentive and decrease the productive and innovative output. It will be hard for a smaller firm to compete initially and attract top talent, unless the intangibles offered are appealing enough and/or the focus of the business is narrow enough. But the large firms spending large sums to develop the technology will certainly have a trickle-down effect to smaller firms over time. And who knows?
    After a few years making millions, the specialists may choose to start a small firm, taking their expertise and applying it to solving social issues or to developing some targeted market solution. Basically, there is a lot of money to be made from developing AI technologies, and the fastest way to develop them is from incentivizing the specialty. Large firms have the resources to spend, beyond what many universities do, and the greatest benefit will be achieved from using the resources to innovate sooner than later.

Leave a Reply

Your email address will not be published. Required fields are marked *