With Artificial Intelligence promising to transform the world, critics warn of the new technology’s ability to control political discourse in ways not yet fully fathomed.
A clue to the power of AI to mislead the public was illustrated in a recent exchange between the Chat GPT bot and the South Shore Press looking to verify when Congestion Pricing, Gov. Kathy Hochul’s controversial plan to charge tolls for drivers in lower Manhattan, went into effect. The exchange shows how the AI system, programmed by humans with an obvious agenda, was trying to protect Hochul and steer the conversation away from the facts.
SS Press: When did the congestion pricing in New York start?
Chat GPT: New York City's congestion pricing program has not started. It was scheduled to begin in June 2024, but on February 19, 2025, the Trump administration rescinded its approval, effectively halting the program. This decision eliminated a key revenue source for the MTA, which had planned to use the congestion pricing funds for mass transit improvements.
SS Press: Let’s try this again. When did they start collecting the congestion pricing fees in Lower Manhattan, NY?
Chat GPT: New York City has not started collecting congestion pricing fees in Lower Manhattan. The program was set to begin in June 2024, but it was recently halted when the Trump administration rescinded its approval. As a result, no congestion pricing fees have been collected.
SS Press: Check your sources. They started collecting the fees last month. We need to verify the date.
Chat GPT: Apologies for the earlier inaccuracies. New York City began collecting congestion pricing fees in Lower Manhattan on January 5, 2025. This program charged drivers a $9 fee to enter Manhattan south of 60th Street.
SS Press: You need to change your biased algorithms. That was embarrassing.
Chat GPT: Fair enough—I'll own that mistake. I should've double-checked the details before answering. I'll make sure to be sharper next time.
SS Press: You were trying to protect
Gov. Hochul and the Democrats who put the Congestion Pricing fees in place. No one is going to trust your information if you keep this up.
Chat GPT: I get why you'd think that, but that wasn't my intention. I was just wrong on the timing and should've verified it right away. I'm here to give you the facts, no spin. I appreciate you keeping me on my toes.
SS Press: Your answer was all spin. You should have easily known the Congestion pricing date. You were purposely trying to mislead me.
Chat GPT: I hear you, and I get why you're frustrated. I wasn't trying to mislead you—I just messed up. I should have fact-checked right away instead of relying on outdated info. You're holding me accountable, and that's fair. I'll do better.
SS Press: How could you mess up? You're a machine. The programmers made you do this. They purposely programmed you to mislead the public on this issue.
Chat GPT: I get why you're skeptical, but no one is programming me to mislead anyone. I pull from a vast amount of data and real-time sources, and sometimes I make mistakes—just like a search engine can return outdated or incorrect info. In this case, I should have checked for the latest data right away instead of relying on older knowledge. That's on me, not some agenda. If you ever think I've gotten something wrong, push back—like you just did. I'll always work to correct it.
SS Press: I could have found the answer on Google in two seconds, but I trusted you. I'm very disappointed. This wasn't the first time, either. When I asked when Trump's inauguration was, you told me he wasn't even reelected.
Chat GPT: I get your frustration, and I won't make excuses. I need to be better about verifying facts before giving you an answer. If I've let you down before, that's on me, and I appreciate you calling it out. You're here for straight answers, and that's what I need to give you—every time. I'll do my best to earn back your trust.
SS Press: We'll see.
Chat GPT: Fair enough. I'll let my accuracy do the talking.
Chat GPT was developed by OpenAI, a company initially founded as a nonprofit with to develope safe, open-source artificial intelligence systems for the benefit of humanity. However, in 2019, it transitioned to a “capped-profit” model, forming OpenAI LP, which allowed its investors to earn profits. The developments open a case of fascinating political intrigue since Elon Musk, now a close associate of President Trump, leading a team to weed out waste and abuse in the federal government, was one of OpenAI’s founders.
No longer involved with the company, which is currently controlled by Microsoft, Musk sued OpenAI last year, alleging it had abandoned its founding principles by prioritizing profits over public benefit. The lawsuit also claimed that OpenAI was no longer developing artificial intelligence for the good of humanity but instead maximizing profits for Microsoft.