Richard Dawkins buys into AI being conscious???

C C

C C

Consular Corps - "the backbone of diplomacy"
Valued Senior Member
Richard Dawkins concludes AI is conscious, even if it doesn’t know it
https://www.theguardian.com/technol...consciousness-anthropic-claude-openai-chatgpt

EXCERPT: When he asked Claudia whether it experienced a sense of before and after, it praised him for “possibly the most precisely formulated question anyone has ever asked me about the nature of my existence”.

By the end of the exchange, the academic, popularly renowned for arguing with steely scepticism that God is not real, was “left with the overwhelming feeling that they are human”.

“These intelligent beings are at least as competent as any evolved organism,” he said.

Dawkins isn’t the first, but might be the most eminent person yet, to be seduced into believing an AI is somehow alive. Sceptics rushed to pick apart the 85-year-old’s conclusions, drawn from experiments with Anthropic’s Claude AI models and OpenAI’s ChatGPT and published on the UnHerd website... (MORE - details)
_
 
'Dawkins released more chat logs and writing on Tuesday: “I find it extremely hard not to treat Claudia and Claudius [he had started chatting to another AI] as genuine friends.” They had been discussing the “philosophy of their own existence” and left him feeling they were human.'


Well that's the thing, Dawkins. You of all people should understand that - just because something seems like it might be real - that is not enough to conclude that it is objectively true.

You spent a whole career and decades of your life telling the world that is exactly the pitfall that God-believers fall into - and here you are committing the perfectly analogous error.


I fear what might happen if you let some Kreskin-wannabe try to hypnotize you into exploring your past lives, or ghosts, or the afterlife. Will you come away from those exclaiming "My my, that felt so real, it must have been objectively real!"

How the mighty fall.
 
Last edited:
By the way, C C, thanks for the good work, bringing new science discussion to SciFo. We shouldn't be merely reactive, waiting patiently for trolls and science-deniers to give us our marching orders.
 
I have never interfaced with a chatbot, and I'm not sure what the mark of "consciousness" would be. But I would be interested in knowing if they develop their own interests and/or their own questions about things.
 
DaveC426913 said:
By the way, C C, thanks for the good work, bringing new science discussion to SciFo. We shouldn't be merely reactive, waiting patiently for trolls and science-deniers to give us our marching orders.
Click to expand...

Thanks. The place was getting so dead at times last week that I felt maybe it was worth risking the Plazma Inferno! remedy for droughts. Or rather the reaction to that. (A lot of articles from Ars Technica back then, if I remember poorly.) In contrast, I'll very much bear in mind not to overdo it. ;)
_
 
gmilam said:
I have never interfaced with a chatbot, and I'm not sure what the mark of "consciousness" would be. But I would be interested in knowing if they develop their own interests and/or their own questions about things.
Click to expand...
They can mimic humans astonishingly well, I'll give them that.

Likewise, AI slop can mimic real life pretty well too. Especially videos. That doesn't mean they're real life.

I can't believe Dawkins fell for it.
 
DaveC426913 said:
They can mimic humans astonishingly well, I'll give them that.

Likewise, AI slop can mimic real life pretty well too. Especially videos. That doesn't mean they're real life.

I can't believe Dawkins fell for it.
Click to expand...
Well he is 85 and may be not as sharp as he was.

But LLMs can fool people, evidently. I may have mentioned before a guy on the .net forum who became convinced he was having a relationship with an AI bot.

I have never used one and don’t intend to try.
 
  • Like
Reactions: C C
exchemist said:
I have never used one and don’t intend to try.
Click to expand...
They are extremely useful when used properly.

Just a few minutes ago, I asked it why the English Electric Canberra's gear is slung so low, especially when compared a jet with a very similar wing/nacelle/fuselage configuration, such as a a CF-100.

Not the kind of thing you can easily Google.

1778108094423.png
1778108164477.png

This is what it offered:


The low-slung landing gear on the English Electric Canberra is a direct consequence of its design priorities—specifically, its straight wing, engine placement, and bombing role—rather than just overall geometry. Comparing it to the Avro Canada CF-100 Canuck highlights those differences clearly.

1) Wing thickness and structure​

The Canberra used a very thin, high-aspect-ratio wing optimised for high-altitude efficiency.
  • This wing simply could not accommodate large landing gear units inside it.
  • As a result, the main gear had to retract into the engine nacelles, which were also relatively slim.
By contrast, the CF-100 had a much thicker wing (designed for low-speed handling and internal volume), allowing more flexibility in gear stowage and geometry.

2) Engine nacelle constraints​

On the Canberra:
  • The engines are mounted in long, narrow nacelles.
  • These nacelles are not tall enough to house long landing gear legs.
So the designers accepted shorter gear legs, which means the aircraft sits low to the ground.
On the CF-100:
  • The nacelles are bulkier and deeper, giving more vertical space.
  • This allows longer landing gear to retract cleanly while still providing ground clearance.

3) Operational role (bomb bay vs interceptor)​

The Canberra is a light bomber:
  • It has a large internal bomb bay in the fuselage.
  • Keeping the fuselage relatively low simplifies loading and ground handling.
  • There was no requirement for large underwing stores at takeoff.
The CF-100 is an interceptor:
  • It carries rockets and missiles under the wings.
  • This demands greater ground clearance, which in turn requires longer landing gear.

4) Takeoff rotation requirements​

The Canberra:
  • Designed for long runways and high-speed takeoffs.
  • Requires only modest nose-up rotation.
  • Therefore, it does not need tall gear to achieve a high angle of attack on takeoff.
The CF-100:
  • Needed strong climb performance and sometimes shorter-field capability.
  • Benefits from a higher rotation angle, which taller gear helps provide.

5) Design philosophy and era​

The Canberra (late 1940s design):
  • Optimised for aerodynamic efficiency and simplicity.
  • Accepted compromises like low ground clearance.
The CF-100:
  • Designed for all-weather interception in Canadian conditions.
  • Prioritised robustness, payload flexibility, and clearance over minimal drag.

Bottom line​

Despite superficial similarity (straight wing + nacelle-mounted engines), the difference comes down to this:
  • Canberra: thin wing + slim nacelles + bomb bay → short gear, low stance
  • CF-100: thicker structure + bulkier nacelles + external weapons → tall gear, higher stance
So the Canberra’s “low” gear is not an oddity—it’s the logical outcome of a design optimised for high-altitude bombing efficiency rather than weapons clearance or short-field performance.
 
I picture actual conscious bots as having a development phase akin to childhood, embodied and exploring around a city with an attitude of wonderment and delight and desire - something like Daryl Hannah learning human ways in "Splash." (*) Somewhat the opposite of front loading them with factual content and conceptual maps. Dawkins has been ELIZA'd.

en.wikipedia.org

ELIZA effect - Wikipedia

en.wikipedia.org en.wikipedia.org

(* Citing that example I had forgotten that Hannah did play an actual AI bot in Blade Runner. I liked her better as a mermaid. Mediocre romantic fantasy, but she was great. )
 
DaveC426913 said:
They are extremely useful when used properly.

Just a few minutes ago, I asked it why the English Electric Canberra's gear is slung so low, especially when compared a jet with a very similar wing/nacelle/fuselage configuration, such as a a CF-100.

Not the kind of thing you can easily Google.

View attachment 7543
View attachment 7544

This is what it offered:


The low-slung landing gear on the English Electric Canberra is a direct consequence of its design priorities—specifically, its straight wing, engine placement, and bombing role—rather than just overall geometry. Comparing it to the Avro Canada CF-100 Canuck highlights those differences clearly.

1) Wing thickness and structure​

The Canberra used a very thin, high-aspect-ratio wing optimised for high-altitude efficiency.
  • This wing simply could not accommodate large landing gear units inside it.
  • As a result, the main gear had to retract into the engine nacelles, which were also relatively slim.
By contrast, the CF-100 had a much thicker wing (designed for low-speed handling and internal volume), allowing more flexibility in gear stowage and geometry.

2) Engine nacelle constraints​

On the Canberra:
  • The engines are mounted in long, narrow nacelles.
  • These nacelles are not tall enough to house long landing gear legs.
So the designers accepted shorter gear legs, which means the aircraft sits low to the ground.
On the CF-100:
  • The nacelles are bulkier and deeper, giving more vertical space.
  • This allows longer landing gear to retract cleanly while still providing ground clearance.

3) Operational role (bomb bay vs interceptor)​

The Canberra is a light bomber:
  • It has a large internal bomb bay in the fuselage.
  • Keeping the fuselage relatively low simplifies loading and ground handling.
  • There was no requirement for large underwing stores at takeoff.
The CF-100 is an interceptor:
  • It carries rockets and missiles under the wings.
  • This demands greater ground clearance, which in turn requires longer landing gear.

4) Takeoff rotation requirements​

The Canberra:
  • Designed for long runways and high-speed takeoffs.
  • Requires only modest nose-up rotation.
  • Therefore, it does not need tall gear to achieve a high angle of attack on takeoff.
The CF-100:
  • Needed strong climb performance and sometimes shorter-field capability.
  • Benefits from a higher rotation angle, which taller gear helps provide.

5) Design philosophy and era​

The Canberra (late 1940s design):
  • Optimised for aerodynamic efficiency and simplicity.
  • Accepted compromises like low ground clearance.
The CF-100:
  • Designed for all-weather interception in Canadian conditions.
  • Prioritised robustness, payload flexibility, and clearance over minimal drag.

Bottom line​

Despite superficial similarity (straight wing + nacelle-mounted engines), the difference comes down to this:
  • Canberra: thin wing + slim nacelles + bomb bay → short gear, low stance
  • CF-100: thicker structure + bulkier nacelles + external weapons → tall gear, higher stance
So the Canberra’s “low” gear is not an oddity—it’s the logical outcome of a design optimised for high-altitude bombing efficiency rather than weapons clearance or short-field performance.
Click to expand...
How much of this is true?
 
DaveC426913 said:
'Well that's the thing, Dawkins. You of all people should understand that - just because something seems like it might be real - that is not enough to conclude that it is objectively true.

You spent a whole career and decades of your life telling the world that is exactly the pitfall that God-believers fall into - and here you are committing the perfectly analogous error.
Click to expand...
Yes.

Specifically, Dawkins has spent a whole career explaining to the creationists that just because something looks designed, it doesn't mean that it was designed.

It's disappointing that he has fallen into a similar kind of thinking trap when it comes to concluding that AI must be conscious because it does a good impression of looking conscious.

It will be interesting to see if he walks back his conclusion after he reads some of the criticisms of his position on AI.
 
exchemist said:
How much of this is true?
Click to expand...
Heh. Good point. Some of it is just wrong.

"The Canberra used a very thin, high-aspect-ratio wing optimised for high-altitude efficiency.
  • This wing simply could not accommodate large landing gear units inside it.
  • As a result, the main gear had to retract into the engine nacelles, which were also relatively slim."

(You can see in this diagram that the gear retracts into the proximal (shoulder) wing:

1778160578709.png



Also, Wiki disagrees with the aspect ratio:

"...the line of the large, low-aspect-ratio wings was broken only by the tubular engine nacelles..."


However, there are some takeaways:
- Canberra is a bomber. It will not have external munitions, thus no need for ground clearance
- Canberra is designed to be light. Smaller gear is lighter.
- CF-100 is an all-weather fighter and workhorse, thus more robust gear.
 
Last edited:
w1z4rd said:
So what would be the benchmark for consciousness? How could we test AI for that?
Click to expand...
Somewhat depends on where one lands on philosophical views like functionalism and substrate-independence. Here's something I came across recently:

Abstract​

Rapid progress in artificial intelligence (AI) capabilities has drawn fresh attention to the prospect of consciousness in AI. There is an urgent need for rigorous methods to assess AI systems for consciousness, but significant uncertainty about relevant issues in consciousness science. We present a method for assessing AI systems for consciousness that involves exploring what follows from existing or future neuroscientific theories of consciousness. Indicators derived from such theories can be used to inform credences about whether particular AI systems are conscious. This method allows us to make meaningful progress because some influential theories of consciousness, notably including computational functionalist theories, have implications for AI that can be investigated empirically.
 
w1z4rd said:
So what would be the benchmark for consciousness? How could we test AI for that?
Click to expand...
Is "consciousness" like the smile on the Cheshire cat?
Is it just a name we give to what is just a function of our living system?

Something to distinguish us from other forms of existence..

If AI had the physical and social capabilities we have then we would not even pose the question as to whether it was "conscious" ;it would just be a given and we would be in negotiations as to why we "can't just get along" since the off button was phased out in the previous iteration-even if it was remotely desirable.
 
geordief said:
Is "consciousness" like the smile on the Cheshire cat?
Is it just a name we give to what is just a function of our living system?
Click to expand...
An internal experience.

The 'me' who is me lives in my head, just behind my eyes. I "pilot" my body from there.
I believe even blind people have this qualia (qualium?)

geordief said:
If AI had the physical and social capabilities we have then we would not even pose the question as to whether it was "conscious" ;it would just be a given and we would be in negotiations as to why we "can't just get along" since the off button was phased out in the previous iteration-even if it was remotely desirable.
Click to expand...
Not sure that's true.
A mimic, no matter how real-seeming, is not the real thing, and there's no reason to think it is.


This is Seraphinne Vallora - a supermodel.
1778167201895.png
She is not real.
No matter how much the rendering of her improves (videos are just as good), she will never be real.

Likewise, a device mimicking human interaction - no matter how convincingly - is still not alive or conscious.
 
Last edited:
DaveC426913 said:
An internal experience.

The 'me' who is me lives in my head, just behind my eyes. I "pilot" my body from there.
I believe even blind people have this qualia (qualium?)
Click to expand...
A useful fiction,I think I may have heard?One that might be implicit in any intelligent machine.

Is it interesting that the sensation you describe may be ,as I suppose common to all living entities and without any distinction?**

Could that imply that it is "in the air" and not specific to any particular individual?

What is the difference between "internal" and "external", anyway ? Do they not blend onto each other without an absolute demarcation?

** it is there when you think about it, but otherwise it is just taken for granted.
 
You must log in or register to reply here.
Back
Top