Can YOU tell the difference between a real person and an AI bot?

Popular AI chatbots like ChatGPT and Bard are designed to mimic human speech as closely as possible.

And as deep learning technology becomes more sophisticated, it becomes difficult to distinguish these computer models from real people.

Now, one free online game gives you two minutes to talk to someone (or something) and guess if they’re a fellow human or an AI.

“Human or not?” was inspired by the Turing test, devised by legendary British computer scientist Alan Turing in 1950.

A computer passes the so-called test when someone cannot properly tell the difference between a human reaction and an AI reaction.

When you start the game, you or your chat partner will be chosen to start the conversation.

You can then talk about anything you want for two minutes, though the participant can leave the conversation before the time runs out, even if it’s an AI.

At the end you have to select who you think you have spoken to: a real person somewhere in the world or an autonomous computer program.

MailOnline had 10 conversations and was able to guess six times whether it was a human or a bot.

But overall, the AI ​​convincingly replicated human speech, making it a tricky but fun puzzle.

Human or not? was created by developers of Tel Aviv-based company AI21 Labs, following the popularity of ChatGPT, released in November.

Both ChatGPT and Bard, which is made by Google, are built using large language models (LLMs) – deep learning algorithms that can recognize and generate text based on knowledge gained from huge data sets.

AI21 creative director Amos Meron told ZDNet that the game was created using the company’s own LLM, called Jurassic-2, as a “social experiment.”

“I’ve had many conversations with friends and colleagues about AI over the past few months,” he said.

“We have a lot of assumptions about how people would interact with AI bots in the near future, and about what is perceived as human behavior online.

‘Here I was thinking about setting up a social experiment that would allow everyone to challenge these assumptions for themselves.

“At a fundamental level, we want people to have an interesting and thought-provoking experience with the game

“We hope it will stimulate a wider public conversation about how we can make better informed, fairer and safer use of AI technology.”

Human or not? is described as a “social Turing game”, referring to the Turing test, introduced by Alan Turing in 1950 as part of his seminal paper Computer machines and intelligence.

A human interrogator asks an AI and another human interrogator asks questions and evaluates the answers – but he or she doesn’t know which is which.

If the AI ​​fools the interrogator into thinking his answers are human-generated, he passes the test.

Meron said that the Turing test “plays perhaps its most important role now,” even though it was created more than 70 years ago.

“It can help us understand where we stand, challenge our assumptions and become the catalyst for asking the right questions and starting some much-needed discussion,” he said.

Human or not? joins a long list of chat-based AI programs released in the past six months.

In addition to ChatGPT and Bard, My AI is built into the social media app Snapchat, YouChat from the American search engine You.com and Ernie Bot from the Chinese company Baidu.

ChatGPT, by far the most popular, has taken the world by storm, reaching over 100 million users just three months after its launch in November.

Its success reportedly caused panic among Google and fears that its days as the world’s number one search engine could soon come to an end.

Google scrambled to find Bard, which got a limited release last month — though it outperformed its rival when the two were tested by MailOnline.