Weizenbaum's ELIZA: A Reflection on AI and Transference

Generated by DALL-E

Sometimes, the simplest creations leave the most profound impacts. This was true for Joseph Weizenbaum's ELIZA, a chatbot I became familiar with during my MSc studies in 1985. My first assignment was to code a version of ELIZA in Prolog, and it was surprisingly easy. Yet, the implications of this simple program were anything but.

ELIZA, created in the mid-1960s, was one of the earliest examples of what we now call a chatbot. Its most famous script, DOCTOR, simulated a Rogerian psychotherapist. This simplicity was deceptive; the program merely echoed user inputs in the form of questions, yet it evoked profound emotional responses from users.

(You can try out ELIZA for yourself here.)

When I was tasked with coding ELIZA in Prolog as a new AI MSc student, I was struck by the simplicity of the task. Prolog, with its natural language processing capabilities, seemed almost tailor-made for this assignment. The ease with which I could replicate aspects of ELIZA's functionality was both exhilarating and unnerving. It was a testament to both the power of declarative AI programming languages like Prolog and the ingenious design of ELIZA.

The real intrigue of ELIZA lies not in its technical complexity but in the psychological phenomenon recognised by Freud it inadvertently uncovered: transference. Users often attributed understanding, empathy, and even human-like concern to ELIZA despite knowing it was a mere program. This phenomenon highlighted the human tendency to anthropomorphise and seek connection, even in unlikely places.

Joseph Weizenbaum himself was startled by this phenomenon. As a technologist who understood the mechanical underpinnings of ELIZA, he was disturbed by the emotional attachment users developed with the program. This led him to become a vocal critic of unrestrained AI development, warning of the ethical and psychological implications.

My journey with ELIZA and Prolog was more than an academic exercise; it was a window into the complex relationship between humans and AI. It highlighted the ease with which we can create seemingly intelligent systems and the profound, often unintended, psychological impacts they can have. As we venture further into the age of ChatGPT, Weizenbaum's cautionary tale remains as relevant as ever.

In an era where AI is more advanced and pervasive, revisiting the lessons from ELIZA and Weizenbaum's reflections, as highlighted in articles like this recent one from The Guardian, is crucial. It reminds us that in our quest to advance AI, we must remain vigilant of the human element at the core of our interactions with machines. Weizenbaum's legacy, through Eliza, is not just a technological artefact but a cautionary tale about the depth of human interaction with machines and the ethical boundaries we must navigate as we move ahead in the realm of AI.

#Turing Test hasn't been passed

Sorry to confuse you, since my previous blog post claimed the Turing Test had been passed. On closer inspection scientists are now not so sure that the chat bot "Eugene Goostman", a computer programme developed to simulate a 13-year-old Ukrainian boy, actually passed Turing's Imitation Game. Let's look at the evidence. First, Eugene only managed to convince 33% of the judges that it was human, therefore the majority of thought it wasn't human. Second, Eugene's developers are clearly gaming the game; by creating a persona who is a child and speaks English as a second language they're lowering the judges' expectations as to its ability to understand questions and give sensible answers. Transcripts of the conversations from the latest Royal Society test are not yet available but here's and excerpt from a transcript from a previous event at Bletchley Park on 23 June 2012, with the same program.
[15:02:56] Judge: I come from Manchester. Did you see the jubilee?
[15:03:02] Eugene: Could you invite me to Manchester? I would be so grateful. 
[15:03:17] Judge: Did you see the jubliee?
[15:03:27] Eugene: Try to guess! Actually, I don't understand why you are interested. I know you are supposed to trick me.
[15:03:35] Judge: Not really
[15:03:40] Eugene: Not really? Why? What are any interesting site-seeings in Manchester?
[15:03:59] Judge: Lots of nice things. Do you like football?
[15:04:04] Eugene: Does anyone hate it? Some another topic?
[15:04:18] Judge: Will you be watching the football on Sunday?

[15:04:23] Eugene: I play sports sometimes. On computer. But I prefer racing.

Let's analyse it. First, Eugene doesn't know what the Jubilee was - perhaps it didn't make the news in Ukraine, that's understandable. But then something odd happens; Manchester and football are mentioned and Eugene tries to change the topic. I'm sorry but there isn't a 13 year old boy on the planet who wouldn't then have mention Manchester United! FAIL. You can make up your own mind by talking with Eugene here (hint: try the Manchester and football trick, Eugene falls for it every time, and yes football is a big sport in Ukraine).

Mitsuku chat bot wins AI's Loebner contest

The BBC has reported that a chat bot called Mitsuku, originally designed for a music website, has won the annual AI Loebner contest. The annual Loebner contest is really a Turing test, where chat bots have to convince human judges that they are human. Briton Steve Worswick, who wrote Mitsuku, won $4,000 meaning the judges found it to be the best chat bot entered. Mitsuku did not win the $100,000 Grand Prize meaning it hadn't convinced the judges it was human. Why not give Mitsuku a try yourself and you can be the judge.

#Turing Test fail!

This is an amusing video from the Cornell University Creative Machine Lab of two AI chat bots talking to each other. Cleverbot, which is the chat bot that both AI avatars are using, won the 2011 Loebner Prize in Artificial Intelligence; the annual Turing Test competition for chat bots. It's worth pointing out though that Cleverbot was judged the best bot in competition, but it did not win the Grand Prix of $100,000 for convincing the judges it was human. You can find out how this chat bot duel, more "AI a AI" than "mano a mano", was implemented on the Creative Machine Lab's website. They also have some very cool other demos.