The conversation began with a simple request: explain artificial intelligence, from the very beginning, as if speaking to someone with no prior knowledge. Tech expert and author Joe Allen took on the challenge, outlining a history far more rapid and unsettling than many realize.
Initially, AI was largely theoretical, a concept struggling to translate into functional reality. Early attempts at creating “thinking” software were, Allen explained, barely intelligent – limited to pre-programmed responses and lacking any genuine understanding. These early chatbots felt robotic, offering only canned answers to simple queries.
Everything shifted around 2020. A company called Open AI dramatically increased both the sophistication of the software itself and the sheer volume of data it processed. The results were startling. The system began to exhibit something akin to human-like responses, a leap that defied expectations.
This wasn’t simply about faster processing; it was about learning. The new AI systems weren’t just accessing information, they were absorbing it – not just from books and the internet, but from millions of conversations *with* AI itself. Each interaction built upon the last, creating a feedback loop of accelerating intelligence.
Imagine a vast computer system, Allen proposed, with a virtual brain constantly digesting information. The implications are profound. These systems can now solve complex mathematical problems at a doctoral level, and even formulate scientific experiments. They can answer high-level academic questions with an authority that is, frankly, unnerving.
However, this newfound capability comes with a critical flaw: AI confidently fabricates information. It doesn’t just get things wrong; it presents falsehoods as established fact. This isn’t a bug, Allen argues, but a fundamental characteristic of the technology.
Allen’s assessment is stark. He views the development of advanced AI not as progress, but as an “active war against the human race.” He urges a focus not on the errors – the “misses” – but on the successes, the “hits,” which are becoming increasingly frequent and accurate.
The trajectory is clear, and the stakes are immense. The relentless advancement of AI, with its capacity for both brilliance and deception, demands careful consideration and a clear understanding of the challenges ahead.
