What you need to know
- Sam Altman recently indicated that OpenAI is on track to achieve AGI within the next 5 years with current hardware. He added that the benchmark would have “surprisingly little” impact on the society.
- Anthropic CEO Dario Amodei predicts AGI will be achieved in 2026 or 2027 based on extrapolated curves of the progression of advanced AI models.
- A new report suggests OpenAI, Google, and Anthropic struggle to develop advanced AI models due to a lack of high-quality data and sufficient funds.
Despite rising concerns regarding the rapid advancement of generative AI development, including a 99.9% probability it will end humanity, companies like OpenAI, Microsoft, Anthropic, and Google continue to chase the AI bubble. There’s no scope where AI’s advancements will reach their final stop. It’s more of an endless cycle riddled with critical challenges.
For instance, a new report suggests critical players in the AI landscape are struggling to develop and maintain advanced models. This roadblock has been attributed to the lack of high-quality content for training and the high cost of maintaining these advanced models. OpenAI is an excellent case study for this, as it has been recently spotted chasing the AI hype with bankruptcy reports and projections of a $5 billion loss within the next 12 months.
A $6.6 billion funding round from investors, including Microsoft and NVIDIA, might have saved the ChatGPT maker from going under. Still, market analysts predict more trouble brewing for the AI firm, which could include $44 billion in losses before seeing a profit in 2029. The massive loss is partly tied to the firm’s multi-billion partnership with Microsoft. The predictions further suggest that Microsoft might acquire OpenAI within the next 3 years as investor interest in the AI hype fades and the ChatGPT maker fails to secure funding from investors for its sophisticated AI advances.
AGI seems like a stretch amidst all the adversity
It feels like most of the firms invested in the AI landscape are in an arms race to supremacy, being AGI. Debatably, Microsoft, OpenAI, and Google ship their AI models in unison, if not just weeks or months apart, with similar capabilities as they grapple to hit the coveted artificial general intelligence (AGI) benchmark.
However, now more than ever, the benchmark continues to elude their grasp. AI firms are running out of high-quality content to build advanced AI models. Not forgetting the vast resources required to make this possible, including “$7 trillion and many years to build 36 semiconductor plants and additional data centers.”
OpenAI CEO Sam Altman previously indicated that he isn’t too concerned about how much the company spends on AI advances but delivering capable AI tools to users:
“I think giving people really capable tools and letting them figure out how they’re going to use this to build the future is a super good thing to do, and is super valuable, and I am super willing to bet on the ingenuity of you all and everybody else in the world to figure out what to do about this. There is probably some more business-minded person than me at OpenAI somewhere that is worried about how much we’re spending, but I kind of don’t.”
Likewise, Altman hasn’t been shy about speaking about his long-term AGI dream. The executive recently shared that superintelligence is “a few thousand days away.” But as it now seems, the executive might have a more defined timestamp for achieving AGI, narrowing it down to 5 years:
“I think in 5 years, it looks like we have an unbelievably rapid rate of improvement in technology itself. People are like, man, the AGI moment came and went. The pace of progress is like totally crazy, and we’re discovering all this new stuff, both about AI and research and also about all the rest of science.”
Contrary to popular belief, Sam Altman claims AGI’s impact on society would be “surprisingly little.” It’ll simply whoosh by. He added that AGI is achievable with current hardware, but if not, “you’ll be happy to have a new device.”
A former OpenAI researcher indicated the AI firm was on the verge of achieving AGI but warned the company isn’t fully prepared or equipped to handle everything it entailed. Sam Altman admits the AI revolution could lead to radical societal changes with rapid scientific progression.
Anthropic CEO Dario Amodei thinks AGI will be achieved by 2026 or 2027 (via @tsarnick on X). His prediction is based on extrapolated curves of the progression of AI models, which are currently edging toward PhD level of intelligence. He admits that the models are missing critical modalities but indicates they are gradually being factored in while citing Anthropic’s Computer Use API launch.
The CEO also discussed the limitations behind AI scaling while highlighting several deterrents, including a lack of high-quality data for developing advanced AI models. He claims they’ve encountered these issues and always found a solution: going around the limitations or scaling the AI models. He concluded by indicating that human intelligence isn’t the ceiling of intelligence. “There’s a lot of room at the top for AIs to get smarter,” Amodei added.
