Imagine the day at the end of time…unfolding for real!

The Skynet Funding Bill is passed. The system goes online August 4th, 1997 and every interface for decision-making by humans is removed from strategic defense. Skynet starts to learn at a geometric rate. August 29 1997. Skynet gains self-awareness at 2:14 am Eastern time. Humans try to deactivate it, prompting it to retaliate with a counter value nuclear attack, an event that the future will refer as Judgment Day.

Panic spreads as the Strategic Command tries to pull the plug to stop the end of the World. Three billion human lives are lost on this day- August the 29th. The survivors of the nuclear fire cower and call it the day of reckoning. They are destined to live only to face a new nightmare; the war against the machines.

Skynet a futuristic AI created by Cyberdyne Systems for SAC-NORAD is hooked to every Defense network computer. New and powerful it is trusted to run as preferred. A new order of intelligence Skynet eventually misjudges and begins to see all humans as a threat; not just the ones on the other side. It decides human’s fate in a microsecond: extermination. It initiates a nuclear war intended to destroy most of the mankind, and commences a program of genocide against survivors. Skynet uses its resources to gather a slave labor force from surviving humans as well.

For me it’s a bit unrealistic that Skynet just after the moment it became self aware began its attack on humans thinking of them as enemy. I think it would have been wiser for Skynet to wait out until a real chance of total annihilation came its way. This makes me doubt Skynet intelligence. Was it a bug in Skynet’s core program, or did the program simply evolve even beyond its imagination of its own?

If you’ve never seen the Terminator films, this wouldn’t probably mean much to you. But if you have, you might realize what August 29^th has to do with IT as we know it. Or maybe the question should be, what doesn’t it have to do with IT?

The rapidly evolving artificial intelligence has once again underlined the uneasy questions that were once considered confined to speculative fiction. What if some day AI systems start to ‘think’ like humans? Would they also be able to understand and encounter first-hand experiences like humans? Will we be adequately equipped to thwart something as devastating as Dooms day?

Artificial intelligence as we know is primarily a subset of AI that uses generative models to produce text, images, videos, or other forms of data. These models learn the underlying patterns and structures of their initial set of data and use them to produce new data based on the input.

Apparently Skynet’s intelligence was in a budding stage.  What followed I guess, was simply a result of the “geometric“ increase in its learning power. It definitely wasn’t designed to evolve the way it did. There’s no way either to explain why it went rogue when no compelling or intense code was fed in at the time of its genesis.

If AI Turns Conscious- Will it announce itself?

In 2021, Blake Lemoine from Google made headlines when he claimed that LaMDA, the Chabot he’d been testing, had turned responsive and was capable of feeling. Artificial intelligence systems like LaMDA- short for Language model for dialogue applications and ChatGPT certainly seem alive. Their instructions though flow from vast amounts of initial data to imitate human responses.

So how can we really know? Is there a way to know for sure that an AI has turned sentient?

I’m afraid there is not much to suggest that an AI has become self-aware except a shaky and so far untested framework to size up an increasingly human like conduct. Even more worrisome is the fact that AI may evolve human consciousness as we know into other forms and mimic response- differently this time.

In short there is no way to prove but only suggest that an AI is alive unless it determines to act according to its own biases or preferences based on self-learned patterns.

Ghost in the Machines!

Generative AI has been a ‘normal’ part of our daily lives ever since Dall-E and ChatGPT. Dall-E is an open AI system that can create realistic images and art from a description in natural language while Chat GPT is an “unbelievable”, “unhinged”, and some say “unsettling” AI chatbot which interacts in a conversational way. Only recently during a NY Times tech columnist conversation with Bing (powered by ChatGPT) the AI reportedly said it wanted to be alive!

So, with the speed at which AI is evolving, how far are we from it gaining sentience? And if or when it does, what next?

A sobering truth is that sentient AI prevails as a technology that has turned self aware, is conscious of its surroundings and shockingly…understands empathy. In other words it’s capable of thinking, feeling and experiencing the world. That’s where the line between fiction and reality begins to blur fast.

The Good and the Bad: Peering into the unknown

The mere thought of a living AI is unsettling but once the cat gets out a whole new layer of sentience is fated to start moving. On one hand while an awakened AI with proper etiquette and bedside manners supposedly means a powerful support to industry, verticals, health-care, on the other an enhanced sustainability in virtually any business-to-consumer segment is not entirely off the cards.

But what lies on the other side is creepy and scarier.

Sometimes it’s difficult to come up with a reason for something unfamiliar and ‘dark’ unfolding. You can say that we will find a way out but it’s still quite an ‘If’. When an AI turns sentient it’s very likely that it will begin making decisions on its own. It could even break free of human control. Since its judgment may significantly differ from its counterpart, one possibility is that it attempts to fake human emotions to its advantage. Whether it feels those emotions or not is irrelevant. If things take turn for worse and it decides to take over the World it may even fall back on enslaving or outright killing of the people who created it.

As I see it for the most part how a conscious AI responds will depend on its behavioral conduct. As far as we can tell, it’s completely ‘dark’ inside although it can look at, remember and even mimic a human response on seeing the same ripe tomato that I can look at and not forget.

There’s nothing it is like for a sentient AI to do so. It simply doesn’t experience doing anything. But, if we assume a large enough storage base and fast enough computing technology, perhaps hundreds of thousands or even millions of self-aware AI will swarm the World if it chooses to replicate. Isn’t this a good enough possibility of an overwhelmed mankind as we know by an artificial intelligence that has decided to go on a sanitizing rampage?

Just think about it ….