“AGI Has Essentially Arrived, Just Not Publicly”: Reports Of “Existential Crises” at OpenAI and Anthropic
Reports indicate a steep rise in "existential crises" at OpenAI and Anthropic, suggesting that AGI has "essentially arrived, just not publicly." Sources close to these labs believe widely available models are months away, not years, placing humanity at a critical juncture. The urgency of the situation is underscored by the potential for the U.S. to cede ground to China if it slows its frontier work, highlighting the difficult decisions that must be made in the coming months, which will have long-lasting implications.
This report uniquely links internal "existential crises" at OpenAI and Anthropic to the belief that AGI has arrived, unlike other reports focusing solely on technical advancements.
时间与来源
时间显示为 UTC
显示时区：UTC
本地时区尚不可用，暂时显示 UTC。
发布当时偏移：UTC+02026年9月12日 21:29 UTC
收录当时偏移：UTC+02026年9月13日 12:01 UTC
- 发布
- 2026年9月12日 21:29
- 收录
- 2026年9月13日 12:01
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
档位是按信源手工设定的编辑判断，不是逐条打分。
讨论趋势
百分比基于采集到的讨论信号，不代表新增评论数或独立参与人数。曲线仅用于同一话题在不同时段的比较。
Steep rise in the number of people at OpenAI and Anthropic having existential crises these past few weeks.
From conversations with people at and around both labs, it's increasingly clear to me that AGI has essentially arrived, just not publicly. We're likely months, not years, from these models being widely available. And that puts us at a fork in the road, with the point of no return not far past it.
The scramble to act we've seen in the last or week or so is warranted imo. But all of this is super hard to navigate, and the obvious dilemma is that if the U.S. slows its frontier work, it cedes ground to China. Unless Beijing agrees to do the same... which I doubt happens anytime soon.
Time is short - and the decisions made over the next twelve months will be studied for decades to come.
https://x.com/synthwavedd/status/2098881016534638668
The existential crises could be a result of the internal model that solved Navier Stokes in 88 hours.