In the coming decades, 2023 will be remembered as the year of the AI explosionwhere ChatGPT became arguably the fastest-embracing new technology in human history, and expectations of AI-driven wealth became commonplace. And the year 2024 will be the time to recalibrate expectations.
Of course AI is an impressive technology, and it provides huge opportunities to improve productivity in a series of tasks. But because the hype has gotten so far ahead of reality, the technology setbacks of 2024 may be equally more memorable.
More evidence will surely emerge that A.I. provides many times false information: AI simply invents and makes mistakes. And hopes for a quick fix to the rambling problem through supervised learning, in which these models are taught to stay away from questionable sources, will prove optimistic at best.
Because the architecture of these models is based on predict the next word, or words in a sequence, it is difficult to make predictions based on known truths. And the promise that there will be exponential improvements in productivity, and much vaunted first steps towards “ultimate artificial intelligence”, they are still just that: promises. There is still work to be done to conclude what human tasks can be augmented with these models, and what types of additional training workers need to make this a reality.
There will be no shortage of those who will begin to recognize that It was always a pipe dream to arrive at something resembling human cognition.. Others will say that ultimate artificial intelligence is just around the corner. And also those who will continue to talk about the “existential risks” of AI, such as its uncontrolled deployment in employment issues, beyond the generation of fake news and the risks to democracy. All of this will be part of the discussions in 2024.
The greatest use of ChatGPT will be seen on social networks and web content. The platforms will continue to monetize the “cheap” information through individualized digital ads, while competition for users’ attention will intensify. And online manipulation and misinformation will grow.
There will eventually be more AI startups, and the open source model will gain traction, but it won’t be enough to stop the duopoly in the industry: with Google and Microsoft/OpenAI dominating the field with its gigantic models. And many more companies will be forced to rely on these models to develop their own applications.
They will intensify calls for regulation. But antitrust measures will probably go nowhere, because neither courts nor politicians will have the courage to try to dismantle the world’s largest technology companies.
