Can ChatGPT pass the Turing Test?

Can ChatGPT pretend to be human? Maybe artificial intelligence isn't so artificial after all...

Can ChatGPT pass the Turing test?

You can trust PC GuideOur team of experts use a combination of independent consumer research, in-depth testing where appropriate – which will be flagged as such, and market analysis when recommending products, software and services. Find out how we test here.

Last Updated on

The Turing test was developed over seven decades ago, by the famous British mathematician and computer scientist Alan Turing. The test is commonly used to measure whether or not artificial intelligence can think and respond like a human being. As ChatGPT increases in popularity, many are wondering: Can ChatGPT pass the Turing test?

ChatGPT is an online AI chatbot, created by OpenAI. Since its release near the end of last year, many people have been taking advantage of the model’s capabilities. And if you are yet to try it out, you are definitely in for a treat.

The AI model has the ability to do almost any task, as long as you prompt it. You can ask it to write essays, solve coding problems, translate texts, and even discuss any topic your heart desires. In fact, the natural language processing platform is recognised as the most powerful of its kind.

Google’s LaMDA model supposedly passed the Turing test last summer. This then sparked discussion about whether or not ChatGPT could also pass. But before we look into that, let’s explore what the Turing Test really is.

What is the Turing Test?

The Turing Test is a popular and well-known test in the world of AI. To put it simply, the Turing Test is a method to test how human-like a machine’s artificial intelligence is.

This is achieved by measuring the machine’s ability to demonstrate intelligent behavior which could deceive the user into thinking it is also human.

Essential AI Tools

Editor’s pick
Only $0.00019 per word!

Content Guardian – AI Content Checker – One-click, Eight Checks

8 Market leading AI Content Checkers in ONE click. The only 8-in-1 AI content detector platform in the world. We integrate with leading AI content detectors to give unparalleled confidence that your content appear to be written by a human.
EXCLUSIVE DEAL 10,000 free bonus credits

Jasper AI

On-brand AI content wherever you create. 100,000+ customers creating real content with Jasper. One AI tool, all the best models.
TRY FOR FREE

WordAI

10x Your Content Output With AI. Key features – No duplicate content, full control, in built AI content checker. Free trial available.
TRY FOR FREE

Copy.ai

Experience the full power of an AI content generator that delivers premium results in seconds. 8 million users enjoy writing blogs 10x faster, effortlessly creating higher converting social media posts or writing more engaging emails. Sign up for a free trial.
TRY FOR FREE

Writesonic

Create SEO-optimized and plagiarism-free content for your blogs, ads, emails, and website 10X faster. Start for free. No credit card required.

A Turing Test normally involves a human assessor and human evaluator who interacts with both the AI and a human in general conversation. If the machine’s responses are able to mimic human-like responses, and the panel of judges cannot distinguish between the machine and human, then the machine has passed the Turing Test. The credentials of an AI chatbot include dialogue management, language translation, and fluency in responding as a human. Therefore, you’d think this should be a perfect demonstration of its core purpose, no? A perfect measure of machine intelligence!

Passing the test itself should, however, not be confused with the confirmation of true intelligence on the machine’s part. Rather it provides a benchmark to evaluate the progress of artificial intelligence research and AI development themselves. Artificial general intelligence (AGI) is a wholly larger challenge. While we are ever approaching that milestone, current artificial intelligence systems are still “narrow” compared to our definition of AGI.

Turing Test on ChatGPT

The question of whether ChatGPT has passed the Turing test is still up for debate. From interacting with the model itself, it seems that ChatGPT does have the ability to pass the Turing Test.

It seems that the AI model has not undergone a proper Turing Test yet. A major issue with trying to conduct the Turing Test on ChatGPT is that the AI likes to specify that it is a language model in conversation. This is obviously a massive giveaway that the machine is not human and can be seen in Time’s conversation with the bot.

Despite this, writers at Less Wrong suggest that this characteristic could be bypassed with a bit of hard wiring on the back end of the model.

However, according to Metaverse Post, ChatGPT has already passed the Turing Test. They state that a data scientist at Buzzfeed confirmed this result by putting the chatbot up against a human from a call center in the Philippines. ChatGPT, in some cases, managed to deceive evaluators. There were moments when they were not able to decipher which response was the bot and which was the human.

Has GPT 4 passed the Turing test?

OpenAI’s latest model has passed numerous exams. However, it is unclear whether or not GPT-4 could pass the Turing Test at this stage. OpenAI hardwires their chatbots to declare they are a large language model before answering difficult questions. This reveal will ultimately invalidate the whole test! If you are able to get your hands and GPT-4 to remove this aspect of the AI, then the model might be able to pass the Turing test.

Can GPT 3 beat the Turing test?

Just like ChatGPT and GPT-4, there are varying views as to whether or not GPT-3 can beat the Turing test. While many claim that no machine has ever passed the Turing test some still argue that GPT-3 would be able to pass. Ultimately, until GPT-3 has been put through a proper test then it will be hard to tell.

Final Thoughts

Passing the Turing Test is a significant milestone that many in the Ai industry would want to reach, however, with ChatGPT, the jury is still out. The artificial intelligence chatbot does show promise though, but in its current state whereby it ‘alerts’ the user that it is, in fact, an ai bot, it would be all to easy to know it wasn’t a human.

Funmi joined PC Guide in November 2022 and has a knowledge of AI apps, gaming and consumer technology.