BlogLine

ChatGPT: Has Artificial Intelligence Finally Defeated Alan Turing? 

1/24/23

chat bot

By Alexia Roney

The year 2022 is the year of AI chatbot. News stories touted Meta’s chatbot Cicero when it defeated human players on the online game Diplomacy without tipping off its computer-generated origin. Media such as Scientific American published articles on AI chatbot LaMDA such as “Google Engineer Claims AI Chatbot Is Sentient: Why That Matters.” To crown the year off, OpenAI launched ChatGPT on November 30, 2022, with free access to the public to beta test its capabilities. Some have used ChatGPT for dad jokes. Others have used it to write school papers, leading the New York Public Schools to ban ChatGPT.  

But what is ChatGPT or, for that matter, an AI chatbot? And, to paraphrase Scientific American, why does it matter? 

A chatbot is a program that can respond to written or spoken statements. Social media is rife with bots. Some are amusing – if one misspells the statute HIPAA on Reddit, a bot will post a reply with the correct spelling. Others, though, are less benign, replicating and thus amplifying false statements across media platforms. But if a human tries to interact with a bot, that person will quickly realize they are dealing with a program. This is the Turing Test, established by computer scientist Alan Turing in 1950. If a computer can converse with a human via typewriter and consistently fool the human, the computer is capable of something like thinking. The goal of the artificial intelligence or “AI” chatbot is to meet the Turing Test.   

This is what ChatGPT is tantalizingly close to achieving.  

If you go to ChatGPT’s website, you can type a question, any question, and ChatGPT will respond. For example, I typed this question: 

And in that explanation lies the answer to why ChatGPT has caught the attention of cybersecurity firms, software firms, pundits, and professionals. Indeed, Microsoft had already invested $1 billion in OpenAI and its ChatGPT program in 2019 and is now rumored to be increasing its investment by a massive $10 billion this year. ChatGPT has the capacity to impact and transform so many industries. This type of software may be able to turn out near perfect statements, briefs, reports, or papers. It may provide predictive insight into negotiations, as Cicero did in the Diplomacy online game.  

But as beneficial as it appears, ChatGPT has already drawn concerns. Is society ready for this level of AI and its potential impact on our economy and workforce? How can we harness the potential of AI while minimizing the risks of misuse or unintended consequences? Stayed tuned as we explore these questions in a series of upcoming posts on AI and its implications for our lives and the world of technology and cybersecurity.

For more information, please contact Alexia Roney, or your local FMG attorney.