Sample Page Title

In a dialogue about threats posed by AI methods, Sam Altman, OpenAI’s CEO and co-founder, has confirmed that the corporate isn’t at the moment coaching GPT-5, the presumed successor to its AI language mannequin GPT-4, launched this March.

Talking at an occasion at MIT, Altman was requested a few current open letter circulated among the many tech world that requested that labs like OpenAI pause growth of AI methods “extra highly effective than GPT-4.” The letter highlighted issues concerning the security of future methods however has been criticized by many within the business, together with numerous signatories. Specialists disagree concerning the nature of the risk posed by AI (is it existential or extra mundane?) in addition to how the business may go about “pausing” growth within the first place.

At MIT, Altman mentioned the letter was “lacking most technical nuance about the place we want the pause” and famous that an earlier model claimed that OpenAI is at the moment coaching GPT-5. “We’re not and received’t for a while,” mentioned Altman. “So in that sense it was type of foolish.”

Nevertheless, simply because OpenAI isn’t engaged on GPT-5 doesn’t imply it’s not increasing the capabilities of GPT-4 — or, as Altman was eager to emphasize, contemplating the security implications of such work. “We’re doing different issues on prime of GPT-4 that I feel have all types of issues of safety which can be vital to handle and have been completely disregarded of the letter,” he mentioned.

You’ll be able to watch a video of the trade beneath:

Altman’s feedback are fascinating — although not essentially due to what they reveal about OpenAI’s future plans. As an alternative, they spotlight a big problem within the debate about AI security: the problem of measuring and monitoring progress. Altman could say that OpenAI isn’t at the moment coaching GPT-5, however that’s not a very significant assertion.

Among the confusion will be attributed to what I name the fallacy of model numbers: the concept that numbered tech updates mirror particular and linear enhancements in functionality. It’s a false impression that’s been nurtured on this planet of shopper tech for years, the place numbers assigned to new telephones or working methods aspire to the rigor of model management however are actually simply advertising instruments. “Effectively after all the iPhone 35 is best than the iPhone 34,” goes the logic of this method. “The quantity is larger ipso facto the cellphone is best.”

Due to the overlap between the worlds of shopper tech and synthetic intelligence, this similar logic is now typically utilized to methods like OpenAI’s language fashions. That is true not solely of the type of hucksters who publish hyperbolic 🤯 Twitter threads 🤯 predicting that superintelligent AI might be right here in a matter of years as a result of the numbers preserve getting larger but additionally of extra knowledgeable and complicated commentators. As a variety of claims made about AI superintelligence are primarily unfalsifiable, these people depend on comparable rhetoric to get their level throughout. They draw obscure graphs with axes labeled “progress” and “time,” plot a line going up and to the proper, and current this uncritically as proof.

This isn’t to dismiss fears about AI security or ignore the truth that these methods are quickly bettering and never totally beneath our management. However it’s to say that there are good arguments and dangerous arguments, and simply because we’ve given a quantity to one thing — be {that a} new cellphone or the idea of intelligence — doesn’t imply now we have the total measure of it.

As an alternative, I feel the main target in these discussions must be on capabilities: on demonstrations of what these methods can and might’t do and predictions of how this will likely change over time.

That’s why Altman’s affirmation that OpenAI isn’t at the moment creating GPT-5 received’t be of any comfort to individuals nervous about AI security. The corporate remains to be increasing the potential of GPT-4 (by connecting it to the web, for instance), and others within the business are constructing equally formidable instruments, letting AI methods act on behalf of customers. There’s additionally all types of labor that’s little question being performed to optimize GPT-4, and OpenAI could launch GPT-4.5 (because it did GPT-3.5) first — one other manner that model numbers can mislead.

Even when the world’s governments have been in some way in a position to implement a ban on new AI developments, it’s clear that society has its palms full with the methods at the moment out there. Certain, GPT-5 isn’t coming but, however does it matter when GPT-4 remains to be not totally understood?