As Sam Altman said, “𝘈𝘐 𝘩𝘢𝘭𝘭𝘶𝘤𝘪𝘯𝘢𝘵𝘦𝘴, 𝘪𝘵 𝘴𝘩𝘰𝘶𝘭𝘥 𝘣𝘦 𝘵𝘩𝘦 𝘵𝘦𝘤𝘩 𝘺𝘰𝘶 𝘥𝘰𝘯’𝘵 𝘵𝘳𝘶𝘴𝘵 𝘵𝘩𝘢𝘵 𝘮𝘶𝘤𝘩.” This reminds us that while AI can generate answers, ideas and even code, it can also make up facts with confidence.
It’s like a genius that sometimes dreams instead of thinks.
We know this, yet we often forget. We get amazed by how well it writes, answers, or codes, and we start trusting it like it’s always right. But the truth is, even when AI sounds confident, it might be giving you a fake source, a made-up fact, or a logic that breaks down when you look closely.
Sometimes, it even argues both sides of a topic without realizing it’s contradicting itself. It can simulate emotions like empathy or urgency, but it’s not feeling anything, it’s just predicting patterns based on data.
We forget that AI can invent fake research papers, confidently tell you something that doesn’t exist, or carry over strange biases hidden in language. These are not rare glitches, they happen more often than we think, just in subtle ways.
What’s even scarier is how much people are starting to trust it without question. The more polished it sounds, the more we assume it must be right. That blind trust is dangerous. When we stop verifying, stop questioning, and let AI take the wheel without oversight, we risk building decisions, opinions and even systems on foundations that might be flawed or completely fictional.
AI can take us far, but only if we stay awake while using it. It’s a guide, not a truth-teller. The goal isn’t to avoid AI, it’s to avoid blindly trusting it.
