ChatGPT on cover of Time Magazine (opens in new tab)

Another example of an article with lots of ifs and coulds, indicating that the authors are just cluelessly jumping on the AI hype bubble like everyone else is for clicks. Quickly analysing a spike in usage isn't the full story. It is just a moment of mania.

It is always what happens afterwards that counts.

   > Stable Diffusion quickly became the talk of the internet. Millions of users were enchanted by its ability to create art seemingly from scratch ...  “It shocked OpenAI and Google, because now the world was able to use tools that they had gated...

Of course it did. That is what you call a permanent and complete disruption of the incumbent(s) out of no where; especially directly at OpenAI's SaaS business plans. Thanks to open-source.

   > If search engines successfully integrate AI, that subtle shift could decimate the many businesses that rely on search, either for ad traffic or business referrals.

Again, 'If' and 'could'. What we've seen with Bing AI and Bard is that they can't be trusted to give us factual results - only hallucinated answers which can't be relied on with little to no references on how the AI got its answer; let alone a transparent explanation of it. Another failed use-case that has collapsed right in front of us.

   > But while AI has been subject to a similar level of breathless hype, the difference is that the technology behind AI is already useful to consumers and getting better at a breakneck pace: AI’s computational power is doubling every six to 10 months, researchers say. It is exactly this immense power that makes the current moment so electrifying—and so dangerous.

Other than summarization, it fails at everything else at the expense of burning the planet to 'improve it'. Even just confusing it with a bad pixel, prompt-injection, font tricks without knowing why it is getting confused it is essentially disqualified in very serious applications; like finance, law, medical and transportation applications.

Like it or not, open-source and transparent AI models are the future. Not short-lived black-box hype experiments like ChatGPT or the unhinged Bing AI who's ancestry is derived from Tay. Both of these 'AIs' will be overtaken by an open-source equivalent.

There is nothing 'revolutionary' about this hype cycle other than 'train it on more data and create another black-box closed AI SaaS'.