Artificial Intelligence: Promises, Challenges, and the Bubble
“A COMPUTER CAN NEVER BE HELD ACCOUNTABLE
THEREFORE A COMPUTER MUST NEVER MAKE A MANAGEMENT DECISION“
IBM slide from 1979
Revisiting the topic of AI, which I previously discussed in the article “Between the Promise and the Danger of AI,” I would like to add some important recent developments.
Challenges and Problems in the AI Industry
Artificial intelligence has been the target of significant investments, with exaggerated promises of revolutionizing various sectors. However, the financial reality of the AI sector is complex and problematic. A recent article on Futurism points out that despite the enthusiasm and resources invested, many companies still fail to turn their AI initiatives into concrete profits. The technology is expensive to develop and implement, and the financial return is not always immediate.
High Cost and Slow Adoption
Developing AI solutions involves significant costs in research, development, and technological infrastructure. Additionally, the large-scale adoption of AI faces critical barriers, such as concerns over data security, privacy, and the reliability of autonomous systems. Many companies are still in the experimentation phase and evaluating the real benefits of AI, which delays the return on investment.
Examples of “Success” and Failure
While some companies, like Nvidia, manage to profit from selling the hardware necessary for AI operations, such as high-performance GPUs (or “success”), most startups and companies focused on AI software face significant difficulties in remaining financially viable. The market is saturated with solutions that promise a lot but deliver results below expectations, leading to growing disillusionment among investors.
Uncertainties and the Possibility of an AI Bubble
Recently, concerns have arisen that we are in an AI bubble, as noted by Robert Armstrong of the Financial Times. If we remove all the major AI stocks from the S&P 500, like Nvidia and other chip companies, along with big tech companies like Apple, Amazon, Alphabet, and Meta, which have risen so much due to AI excitement, the rest of the stock index is actually down over the last three months.
One might argue that this is natural, as AI is seen as the great technology of the future. However, it is important to remember that there are two types of bubbles. The first forms around an idea or technology whose value is vastly overblown, like tulips or cryptocurrencies. The second forms around something genuinely valuable but occurs before the business around that idea or technology has had a chance to fully form. It’s not about questioning the value of AI as a technology but recognizing that we still don’t know who the winners and losers will be in this sector. We are indeed in an AI bubble.
Additional Considerations
Edward Snowden
Edward Snowden recently harshly criticized OpenAI, stating that the company performed a “calculated betrayal of the rights of every person on earth.” Snowden highlighted that the appointment of former NSA Director Paul Nakasone to OpenAI’s board is a move that should not be trusted. “They’ve gone full mask off: do not ever trust OpenAI or its products,” Snowden said. He warned that this decision represents a deliberate betrayal of everyone’s rights, endangering global privacy and security.
Google Researchers
Google researchers published a paper warning that generative AI is ruining vast areas of the internet with fake content, which is especially ironic coming from a company pushing this very technology to its users. The study revealed that most generative AI users harness the technology to “blur the lines between authenticity and deception,” posting fake or manipulated content online. This practice is damaging the collective understanding of socio-political reality and scientific consensus.
The researchers concluded that the mass production of low-quality synthetic content is increasing skepticism towards digital information and overloading users with verification tasks. Furthermore, the problem is exacerbated by Google itself, which has allowed this fake content to proliferate.
Conclusion
Given the many problems and challenges presented, the promise of artificial intelligence seems more like an illusion than a tangible reality. Financial concerns, slow adoption, significant failures, and the risk of being in a bubble indicate a bleak future for AI. The recent appointment of controversial figures to lead AI companies and the questionable practices of tech giants only increase skepticism. Instead of blindly trusting the promises of a technological revolution, it is crucial to adopt a critical and vigilant stance. The future of AI is filled with uncertainties, and the risk that this technology will cause more harm than good cannot be ignored.
For more details, check out the full articles on Futurism: Snowden Calls OpenAI ‘Calculated Betrayal’ and Google Researchers Warn About AI on the Internet, IBM Slide and the video on Financial Times.