I recently needed to contact the CEO of a startup called Lindy, a company developing personal assistants powered by artificial intelligence. Instead of looking for it myself, I turned to an AI helper of my own, an open source program called Auto-GPT, typing in “Find me the email address of the CEO of Lindy AI.”
Like a delightfully enthusiastic intern, Auto-GPT began furiously Googling and browsing the web for answers, providing a running commentary designed to explain its actions as it went. “A web search is a good starting point to gather information about the CEO and their email address,” it told me.
Auto-GPT mentioned that there are multiple sources stating that Flo Crivello is the CEO of Lindy.ai. However, the email address for Flo Crivello has not been found yet. The program will now examine Flo Crivello’s LinkedIn profile in search of their email address. Unfortunately, this attempt was unsuccessful as well. As an alternative, the program proposed the possibility of guessing Crivello’s email address by considering commonly used formats.
Once I granted permission, Auto-GPT utilized various email verification services available on the internet to determine the validity of its guesses. Unfortunately, none of them yielded a definitive response. However, the program saved these addresses in a file on my computer, hinting that I could consider sending emails to all of them.
Who am I to question a friendly chatbot? I tried them all, but every email bounced back. Eventually, I made my own guess at Crivello’s email address based on past experience, and I got it right the first time.
Auto-GPT failed me, but it got close enough to illustrate a coming shift in how we use computers and the web. The ability of bots like ChatGPT to answer an incredible variety of questions means they can correctly describe how to perform a wide range of sophisticated tasks. Connect that with software that can put those descriptions into action and you have an AI helper that can get a lot done.
Certainly, similar to ChatGPT, there will be instances where agents constructed in this manner may generate perplexing messages. As mentioned earlier, although searching for an email address poses minimal risk, there is a possibility that agents in the future may be assigned more daring tasks such as making flight reservations or reaching out to individuals on your behalf. Ensuring that these agents are both intelligent and secure is a significant focus for projects and companies involved in advancing the ChatGPT era.
When I eventually had a conversation with Crivello from Lindy, he displayed a strong belief that AI agents will have the capability to completely replace certain office workers, like executive assistants. He imagines that numerous occupations will just vanish.
