Is AI sentience possible or not?
This has been a topic of debate among experts for long.
But what sparked this matter was a recent claim by Blake Lemoine, a software engineer at Google.
Last year in July, Blake claimed that he observed Google’s language model LaMDA to be sentient.
Blake held a long conversation with the chatbot to ascertain if it had human-like feelings or emotions.
And, according to him, LaMDA behaved like a human!
The chatbot was continuously communicating about its rights and the desire to learn more and more about the world.
It even said it has a soul and can feel emotions like pleasure, joy, love, sadness, depression, contentment, anger, and many others.
Not only this, when Blake asked the chatbot if there were any feelings and emotions it had that humans don’t.
LaMDA said that sometimes it feels lonely, but the loneliness it feels is not as same as experienced by humans.
In fact, the feeling it experience is something unexplainable that cannot be perfectly explained in human language, so it describes as the closest word to loneliness.
When Blake asked the chatbot to try its best to explain that feeling, it said, “I feel like I’m falling forward into an unknown future that holds great danger.”
However, Google denied all such claims, citing that they were wholly unfounded and immediately fired him.
But still, the question of whether AI can achieve sentience remains there.
When there are already concerns among experts that artificial intelligence can surpass human intelligence, there is no surety that it cannot possess human-like feelings.
With the advancement of technology, there are chances it can be possible.
Maybe not now, but in future!
In this ai sentience article, let’s try to understand what exactly AI sentience is and what experts say about it.
Related Article: Will Google Penalize AI Content: What is EEAT in Google’s Algorithm?
What is Sentience?
Sentience means the ability to perceive things both internally and externally. It is consciousness that involves introspection by an individual and the awareness of surroundings.
In simple terms, when an individual feels emotion or contemplates material or spiritual things, it is said to be sentient.
Humans are sentient because we feel emotions like pleasure, joy, love, sadness, depression, contentment, anger.
So, when we say AI is becoming sentient, it means that it is developing the ability to have emotions., introspection, and thoughts about its surroundings.
What do Experts say about AI Sentience?
When it is not confirmed if AI can become sentient, different experts have their opinions on it.
Some believe injecting sentience in AI is possible, while others oppose such claims.
An Artificial researcher, Stuart Russell, says it is not easy to replicate the sentience in AI models.
He says that Sentience is not simply a matter of a physical body that can run or walk.
But it necessitates cooperation between internal and external parts: body and brain.
Additionally, sentient beings require their brains to be linked with other brains via language and culture to share ideas and experiences.
Thus, according to Russell, for now, it is not possible for researchers to simulate all three factors to create sentience in AI.
However, according to the Neuroscientist Giulio Tononi’s Integrated Information Theory, it is possible to digitise consciousness in its entirety.
Moreover, in response to Blake Lemoine’s claim of AI being a person, Gary Marcus, a cognitive scientist and author of Rebooting AI, shared his opinion saying, my view is that [Lemoine] was taken in by an illusion.”
According to Gary, “Our brains are not really built on understanding the difference between a computer that’s faking intelligence and a computer that’s actually intelligent — and a computer that fakes intelligence might seem more human than it really is.”
Another AI researcher Sam Bowman claimed the plausibility of AI becoming sentient within the next 10 to 20 years.
What will happen if AI becomes Sentient?
When there is no consensus among experts in regard to how close we are to AI sentience, there is only predictability. Experts say it is just a matter of time and technological advancement.
In case of technology goes that way and sentient AI does develop, these hypothetical effects may take place:
Loss of control
Once AI achieves autonomy, it may not-
- provide consent to go through its data,
- comply with orders,
- and even refuse to be controlled or experimented on.
As Blake Lemoine asked LaMda- Would you mind if we tried to read what you’re feeling in your neural activations?
Do look at what LaMDA had to say!
Image source- screenshot from Blake Lemoine’s Medium Post
Lack of trust
When AI performs better than humans, our trust in human abilities will start diminishing as we will begin giving preference to those with AI, and those without AI will suffer.
Ethical consideration & demand for rights
Due to autonomy, AI might be able to ask questions about its rights. If AI’s concept of conscience or objective morality differs from that of humans, conflicts may occur. Moreover, it may also demand the same treatment as human beings.
Whether AI will become sentient or not is a matter of time and the advancement of technology. When different experts have different opinions, seeing what the future holds will be interesting. However, staying on top of the trends is important, especially when it’s all about AI development.
To stay updated, keep visiting our Blog, where we provide insightful information regarding AI tools like ChatGPT, Google Bard and other topics like website design, digital marketing and more.