Navigating the AI Winter: Lessons from the Past

During an AI winter, the technology becomes little more than a dirty word

Navigating the AI Winter: As we look back, we observe that the last decade has been an excellent phase for artificial intelligence and its researchers. This was when we saw AI in a more dynamic and evolutionary role, whether in professional or personal lives. Enthusiasm and optimism about AI have generally increased since its low point in the early 1990s. Beginning in 2012, interest in artificial intelligence (especially the sub-field of machine learning) from the research and corporate communities led to a dramatic increase in funding and investment, leading to the current (as of 2023) AI boom.

What Exactly Is AI Winter?

AI is cyclic when it comes to hype and advancement. At a certain point, the technology is reaching dizzying levels of media attention and industry funding. At the same time, it experiences the other side of the AI hype cycle at the next cyclic turn – AI winter. ThinkAutomation states, "These are the times where 'artificial intelligence' wanes in both favor and furtherance. During an AI winter, the technology becomes little more than a dirty word, synonymous with false promises. It steps out of the spotlight, away from the disillusioned eyes of the public." Notably, it is a term that depicts the lowest points in AI where interests are reduced, advancements are inhibited, and funding becomes low. And there is a dense possibility that the world at the current stage is heading toward such a situation. 

What Experts Said on AI Winter

Katja Hoffman, a principal researcher at Microsoft Research in Cambridge, said, "I sense that AI is transitioning to a new phase." However, some experts believe it would be wrong to consider this new phase as AI winter, provided that billions are invested in AI, and there will likely be more breakthroughs ahead. Moreover, Robot Wars Judge Noel Sharkey, a professor of AI and robotics at Sheffield University, told the BBC that he likes the term "AI autumn" – and several others agree. Facebook's former AI director Yann LeCun told IEEE Spectrum, "AI has gone through several AI winters because people claimed things they couldn't deliver." According to AI researcher Filip Piekniewski, who is also a critic, said, "Fears of an impending winter are hardly skin deep. Deep learning has slowed in recent years." In one of his articles in 2018, Piekniewski, after discussing most of the aspects of AI winter and its hype, concluded that "Predicting the AI winter is like predicting a stock market crash – impossible to tell precisely when it happens, but almost certain that it will look at some point. Much like before a stock market crash, there are signs of the impending collapse, but the narrative is so strong that it is effortless to ignore them, even if they are in plain sight. In my opinion, there are such signs of a huge decline in deep learning (and probably in AI in general, as this term has been abused 'ad nausea by corporate propaganda) already visible. Visible in plain sight, yet hidden from the majority by an increasingly intense narrative. How "deep" will this winter be? I have no idea. What will come next? I have no idea. But I'm fairly positive it is coming, perhaps sooner rather than later." 

Conclusion

Call it AI winter or AI autumn; the truth somewhere lies with the fact that what was promised with the hype hasn't been delivered with time, and what will happen next in terms of interest, funding, and implementation is all that can be predicted. An ex-Amazon AI researcher Catherine Breslin said, "In the next decade, I hope we'll see a more measured, realistic view of AI's capability, rather than the hype we've seen so far." Impeding its hype and coming out from the stereotype of 'buzzword,' some hope that AI may work for what it has been programmed. Researchers may cherish what it has to cater to today's reality more than what it could be hyped to do. Samim Winiger, a former AI researcher at Google in Berlin, said, "The manifold of things which were lumped into the term "AI" will be recognized and discussed separately."