You’d think it would be difficult to find something that Stephen Hawking, Elon Musk, Tim Cook, Andrew Ng, and Vladimir Putin all agree on, but here’s one: All agree that artificial intelligence (AI) comes with inherent risks that we know of, as well as a few risks we haven’t fathomed yet. That said, they all agree that it comes with “colossal opportunities.”

Here’s how an AI-enabled process typically works. A person identifies an objective, the AI-empowered machines reach out to all available knowledge sources to perceive truths and conditions related to that objective, and then they take action to achieve that goal. Whether that action is to “inform” or “do” is set in advance, and that’s where everyone becomes concerned, because:

• AI can make mistakes
• Human programmers can make mistakes
• Human programmers can be bad actors

Regarding that first item on the list, we need to look no further than IBM’s Watson. Once considered the epitome of AI, just a few years ago, in spite of digesting troves of medical information, Watson developed a troubling tendency of occasionally issuing faulty and dangerous clinical guidance and directives to doctors for patient cancer treatments. Fortunately, doctors recognized the mistakes.

So far so good. Humans were still in charge.

If AI Runs the Show

But at what point will humans stop being in charge of the tasks we assign to smarter and smarter versions of AI? Regarding the second and third concern noted above, outside of science fiction, it certainly is plausible that an inept or devious programmer could bridge the chasm between AI as an information source and AI as a human substitute taking action, transitioning to a more dangerous “do” outcome, rather than “inform.”

Cyber AI terrorists will undoubtedly learn to leverage this trick.

One such unfortunate scenario would be a programmer casually instructing AI to eradicate cancer. If that human programmer didn’t establish appropriate parameters, the AI might determine that the optimal way to eliminate cancer would be to eradicate anyone with cancerous cells in their body. Since everyone has cancer cells in their body, this would instantly turn into a mass extinction event.

However, if AI is granted or seizes the ability to access the Internet of things (IoT) in order to analyze patient data in real-time, the AI may shift from the “inform” to the “do” instruction and deploy that strategy, pulling plugs and worse.

That seems extreme, yes. But smarter people than me are voicing similar concerns. The end game according to many deep thinkers is that AI could cause the end of mankind or the world as we know it.

If AI is given control

We’ve had some funny discussions recently about someone appearing before a judge, and the person on trial uses the defense, “It wasn’t me, my AI did it!”

• “My AI decided on its own to blackmail 5,000 people yesterday!”
• “My AI decided on its own to try hacking into the Pentagon, or the IRS, or the Russian Embassy!”
• “My AI decided on its own to alter the Harvard, Stanford, or MIT grading system!”
• “My AI decided on its own to set up a bank account in the Cayman Islands!”

After all, the metaverse is a place where we can create idealized avatars of ourselves, and it may be just a matter of time until the lines between the real universe and the metaverse are blurred so that it will be difficult to tell who’s who.

Can it Happen?

While these scenarios sound far-fetched and even a bit humorous, we certainly need to determine if AI advances are setting us on the track of an autonomous and dangerous version of AI.

That was the topic of discussion on the Futurati Podcast interview we had with Dr. Roman Yampolskiy. Dr. Yamplskiy, a tenured professor of Computer Science at the University of Louisville and Director of the Cyber Security Lab, is convinced that as long as we keep developing AI, it will be just a matter of time before it takes over. All attempts to “box it in” or insert other ameliorating strategies can only buy time until we face the inevitable.

A recent study published in the Journal of Artificial Intelligence Research regarding the plausibility of catastrophic outcomes from superintelligent, AI-enabled technology came to a similar conclusion. The authors said it would be impossible to contain or restrain these increasingly smarter AI processes, as that would go against the very nature of supercomputing.

And Manuel Cebrian, one of the co-authors of that study, pointed out that we’ve already strayed down that path: “There are already machines that carry out important tasks independently without the programmers fully understanding how they learned it, a situation that could at some point be uncontrollable and dangerous for humanity.”

Other observers liken this inevitability to the more pessimistic views about climate change. Some people feel we may already be at the point where any steps we can reasonably take to reduce CO2 emissions will at best only delay a cataclysmic ecological reality, given the damage so far and our momentum in that direction.

Can We Stop or Significantly Delay an AI Takeover?

Just like with global warming, with governments working collectively to address that threat through the United Nations, we could possibly see collective action around to prevent an AI cataclysmic event. Conceivably countries could act collectively to put rules and safeguards in place, but if UN’s efforts on global warming are any indication, we’re likely to see uneven compliance across countries and the application of game theories where malevolent countries rely on complying countries to do the right thing to their competitive disadvantage.

Thus, given the tremendous benefits AI technology can provide a nation, combined with the presence of nations and authoritarian states that tend to refuse to be bound by international standards, it’s not at all certain we could develop any reasonably enforceable worldwide limitations on AI applications. And like with every new strain of COVID, bad things are rarely contained by borders.

Most nations on earth have signed on to the OECD’s nonbinding, recommended Principles on Artificial Intelligence that seem to focus on the collective good of AI while acknowledging that related risks should continually be explored. At least it’s a start.

Similarly, there seem to be a number of professional organizations for AI experts. But these, too, apparently are more engaged in sharing best practices than setting enforceable guidelines and professional standards. Hopefully, leaders in these organizations will coalesce around a code of conduct to place appropriate guardrails regarding the exploitation of AI. But all it takes is one bad actor – a criminal or a scientist compelled by fame, curiosity, or their government – to upset the apple cart in spite of the best intentions of 99.99% of AI computer scientists.

How Might it All End?

It’s not easy for a futurist to speculate about the end of the future.

Yes, AI will eventually break down, and yes we may have the ability to turn off the power and disconnect communication lines, but the problem is far bigger than those last-ditch efforts.

The tipping point will be when AI is no longer a tool used for research and specific, limited process improvement. It will happen when AI evolves into a general-purpose, life-enhancing strategy that’s given carte blanche to solve a pressing global problem, manage a wide swath of our lives, or improve an area of national priority: national security, climate change, natural resource allotment, wealth distribution, and just about any major vexing problem or challenge we face.

It will happen as soon as we say once too often, and in the wrong context, that “a machine can do this better than a human.”

That is what will open an AI Pandora’s Box that we won’t be able to close.

How do we evaluate and communicate the evolving capabilities of artificial intelligence (AI)? One way to do it is by describing how closely AI matches up with “real” human intelligence.

We can’t seem to get enough of creating “head-to-head” human matchups with AI in games. Pitting our human champions against AI makes for great headlines and human-interest stories because it illustrates the capabilities of AI in ways that most people can easily relate to.

In the last 75 years, we’ve had some memorable moments where computers with increasingly sophisticated AI prevailed in some very complex games, humbling world champions in the process.

1. Tic-Tac-Toe, 1952

In his Ph.D. dissertation on Human-Computer Interaction, University of Cambridge doctoral candidate A.S. Douglas programmed a room-sized EMPSAC computer mainframe to flawlessly compete in the game of “Noughts and Crosses.” There’s no record of the name of its first human “victim,” but whoever it was, they had to enter their move using digital telephonic technology.

2. Backgammon, 1979

A world-class chess player and AI/computer science instructor at Carnegie Mellon University, Hans Berliner was focused on developing an AI chess program. As an interim step, though, he devised BKG 9.8, a backgammon program that defeated Luigi Villa, the world champion, by a score of 7-1.

This was the first time a computer defeated a world champion in any gaming competition. Two years later, Berliner predicted that a chess AI program would defeat the world chess champion by 1990. He was off by just seven years.

3. Checkers, 1994

Marion Tinsley was a Ph.D. mathematician and university math professor at Florida A&M and Florida State … as well as the former Checkers world champion. In 1992, he initially defeated Chinook, a checkers AI program devised by researchers at Alberta University in Canada.

Two years later, they matched up again, but Dr. Tinsley withdrew after several games for health reasons. He was diagnosed with cancer shortly thereafter. Given those circumstances, it’s a bit callous to call this a loss but regardless, Chinook continued to improve from that point on. It ultimately was perfected, meaning that as long as the human player made no mistakes, every match would end in a draw.

4. Chess, 1997

For many of us, chess was long considered the epitome of gaming intellect and strategy. International championships were covered by mainstream media. World champs like Boris Spassky, Anatoly Karpov, and Bobby Fischer were household names.

Computer scientists, including Hans Berliner of backgammon programming fame, had worked on chess-playing computer programs since the 1950s. In 1989, IBM assembled an all-star team to develop its AI chess program, known as Deep Blue. In its first matchup with the world champion in 1996, Garry Kasparov prevailed. But in their second match a year later, Deep Blue came out on top in their six-game match 2-1 with three draws.

5. Scrabble, 2007

In 2007, the AI Scrabble program, Quackle, defeated former Scrabble World champion, David Boys. To qualify to play Boys, Quackle had to defeat another AI Scrabble program, Maven. At the time, and subsequently, Quackle was used by the media during world-class Scrabble tournaments to describe possible plays confronting the competitors.

6. Jeopardy, 2011

Two of Jeopardy’s most successful contestants to date, Ken Jennings and Brad Rutter, agreed to play a three-game exhibition match against IBM’s Watson – a room-sized computer housed in another location due to its noisy operations and need for a cool environment. The humans had their moments, but in the end, Watson was victorious in spite of the fact that Jeopardy “answers” are known for their subtlety and wordplay.

7. Go, 2016

While its rules are simple, the Chinese game of Go is extremely complex. It’s said that the number of possible moves in a game is greater than the number of atoms in the universe. Contributing to the complexity is the fact that as a Go game progresses, the number of stones and possible moves on the board increases rather than decreases, as in the case of board games like checkers, chess, Scrabble, and others. In spite of these hurdles, in 2016, DeepMind, a UK company purchased by Google two years earlier, defeated Go world champion Lee Sodol in a five-game match.

8. Texas Hold ‘Em, 2017

One year later, the Libratus AI system took on four of the world’s top poker players over a three-week period in “Heads-Up No-Limit Texas Hold ‘Em” … and beat them soundly. Developed by researchers at Carnegie Mellon University, Libratus represented a major step forward in AI learning in an environment of imperfect-information game solving and which requires players to take a long-term strategy vs. seizing on short-term wins and losses.

Why do It?

AI is proving its capabilities more and more each year in the sciences, military, transportation, and business functions like Human Resources and e-commerce. One would think we’ve gone beyond the day when we ask AI to “prove” itself to us by beating us in board or computer games.

However, these exhibitions serve an important purpose. They define the capabilities, and hint at the possible perils of machine thinking and AI decision-making in ways the average person can relate to.

Knowing that AI is powering stock picks and selecting job candidates might be uninspiring to many. But when Ken Jennings, tongue-in-cheek, appends the comment, “I, for one, welcome our new computer overlords” to his Final Jeopardy answer during the Watson exhibition; it put AI into a relatable context for us all.

Beyond Games

That’s why AI developers will and should continue to conduct these kinds of real-life demonstrations. I wrote last year about a recent demonstration of the human-like capabilities of the GPT-3 language prediction tool. This AI program made headlines by drafting essays that read and sounded almost human-like.

But as we said at the time, AI has a long way to go. GPT-3 requires inputs and prompts. It’s a language predictor that determines, usually correctly, the thought or statement that should logically come next given what it’s presented with.

And like all AI, GPT-3 is an example of GOFAI, “good old-fashioned artificial intelligence.” We haven’t progressed much beyond these systems that are based on programmed learning and representational information about the world, rather than intelligently perceived information – wisdom that’s accumulated just like the human brain would gather and process it.

In that sense, AI will still remain artificial until it can defeat a human being in the Game of Life – and I don’t mean the Hasbro version!