What is the AI Winter?

A New Kind of AI Autumn is Coming

Artificial intelligence has a lot of hype in business, and is also being used in controversial ways such as by the Chinese State with facial recognition and data capture as championed by Silicon Valley ad firms like Google and Facebook. All is not okay.

Machine-enabled health care may bring us many benefits in the years to come, but those will be contingent on the ways in which it’s used. AI isn’t just not progressing as fast as we thought it might. The ethics and lack of regulation in its development are downright dangerous.

The last decade was a big one for artificial intelligence but researchers in the field believe that the industry is about to enter a new phase. Deep learning is not a strong path for AI’s development. New regulatory efforts barely touch on the ethics and guidelines required to keep humanity safe from its favorite business obsession.

Not only that, national competition in an AI arms race is getting out of hand. An AI dark age might combine with an AI winter in its progress. But AI has not progressed as much in the last 50 years as we’d like to believe.

Many early ideas about thinking machines appeared in the late 1940s to ’50s by people like Turing or Von Neumann. Turing tried to frame the question of “Can machines think?” differently and created the imitation game, now famously called the Turing Test.

AI research gained much funding from U.S. Defense establishments (ONR and ARPA, later called DARPA) in the hope that these technologies would be useful. But we have spawned an internet of algorithms, harmful to humans in many ways without checks and balances. We have harnessed a Big Data age that has created technological corporations that will only become more powerful to the detriment of people.

Research in AI has its ups and downs. The peaks are known as AI summers, and the troughs AI winters. The 2010s were arguably the hottest AI summer on record, with tech giants repeatedly touting AI’s abilities. However, even with more cash infusion, AI’s actual development might be slowing down again in the early 2020s.

AI pioneer Yoshua Bengio, sometimes called one of the “godfathers of AI”, told the BBC that AI’s abilities were somewhat overhyped in the 2010s by certain companies with an interest in doing so.

Companies like Google, Baidu, ByteDance and Facebook tout their AI as the key to their business strategy, but have these companies truly created good for humanity? AI can learn the fundamentals of quantum physics, yet in the business world AI’s development promotes wealth inequality by the creation of a new kind of monopoly. These tech firms that win the AI race become too powerful for their own good.

Given the billions being invested in AI and the fact that there are likely to be more breakthroughs ahead, some researchers believe it would be wrong to call this new phase an AI winter.

However, the lack of ethical regulation and the weaponization of data against citizens is a kind of dark age, winter, and dystopia that could easily get out of control. AI in today’s world is being misused, sometimes against our human rights.

AI is Not as Advanced as We Think

Early experiments served as an inspiration to create the Dartmouth Summer Project in 1956, where the term AI was coined. Decades later, what AI will become by 2050 might be an existential threat to our own humanity, where we become enslaved by our own technologies and the products these powerful companies create.

An AI winter can take many forms, not just a technological lapse in progress. At the start of the 2010s, one of the world leaders in AI, DeepMind, often referred to something called AGI or “artificial general intelligence” being developed at some point in the future. But realistically AGI appears perhaps centuries, not decades, away.

After the increases in funding and enthusiasm for machine translation in the 1950s and early 1960s, progress stalled. Hutchins called the period of 1967 to 1976 the quiet decade of machine translation.

In the 2020s we may be entering another “AI Autumn”-like plateau period. NLP (personal assistant), autonomous-vehicles and AI in healthcare will also require years of training to get good.

AI’s Hype is Creating a More Dangerous World

DeepMind’s lofty AGI ambitions caught the attention of Google, who paid around £400m for the London-based AI lab in 2014 when it had the following mission statement splashed across its website: “Solve intelligence, and then use that to solve everything else.”

Google’s gobbling up talent and using its machine intelligence for nefarious ends became one of the trends of the 2010s. Near the end of the last decade China replicated this with a whole slew of facial recognition startups that now will dominate the future.

You cannot say that this is the start of an AI Summer, the hype and the dangers are beginning to collide and we are starting to become wary. In 2014, Nick Bostrom, a philosopher at Oxford University, went one step further with his book Superintelligence.

It predicts a world where machines are firmly in control. Dystopia caused by AI is far more probable than imminent extinction from climate change. Humanity seems to have a poor sense of the dangers of its own business inventions.

The invention of the internet, for instance, has not been properly regulated. Instead of porn, misinformation, Ads and echo bubbles, we could have had internet for good. Silicon Valley thought otherwise and our entire evolution around the internet, screens and AI is following a darker path because of it.

We Are Nearing an AI Plateau that Could Be an AI Autumn

The arrival of the AI Autumn will mean a new period of questioning what we have created. Gary Marcus, an AI researcher at New York University, said: “By the end of the decade there was a growing realisation that current techniques can only carry us so far.”

We need AI Winters to help reframe AI’s development and create a world that better utilizes these tools, instead of AI being weaponized for profit and controlled by nefarious corporate and authoritarian actors.

We cannot trust Facebook, the Chinese Government, Google or even the U.S. Government to do the right thing with regards to AI’s development. Their vested interests prevent them from doing so.

“There is a general feeling of plateau,” said Verena Rieser, a professor in conversational AI at Edinburgh’s Herriot Watt University. One AI researcher who wishes to remain anonymous said (to the BBC) that we’re entering a period where we are especially skeptical about AGI. AGI? We cannot even regulate our algorithmic world of feeds, screens and a dirty internet that has less utility, more distraction, and a lot of noise that could be considered a bit evil.

In the history of artificial intelligence, an AI winter is a period of reduced funding and interest in artificial intelligence research. The term was coined by analogy to the idea of a nuclear winter.

An AI winter can also be an ethical and regulatory lapse in how AI is being implemented in society, in systems, on the internet, and against people. It’s rather likely the 2020s will be some of the most profound periods for how the future of AI evolves, for good or evil.

There’s something cryptic in how AI might evolve in the 21st century that could be toxic for future generations that’s not being taken seriously.

Climate change is not the only future war, there’s a war for our future freedoms in relation to an AI-centric technological state that is coming. Powerful corporate and Government interests will inhibit our personal and individual rights and liberties in the future. Are you ready?