Title: 马斯克，最新预测 | BestBlogs.dev
URL Source: https://www.bestblogs.dev/article/17d98c2164?amp%3Butm_medium=feed&%3Butm_campaign=resources&%3Bentry=rss_article_item
Published Time: 2026-07-25 23:52:00
Markdown Content: 76
马斯克，最新预测
In an interview with The Economist, Musk predicted that AI could surpass human intelligence within five years, acknowledged the risks but argued development should not be halted, and suggested industry cooperation to reduce risk. 环 环球时报
Yesterday 694 words (about 3 min) View Source →
Sign in to highlight text and take notes as you read. Sign in now
原创 孔尔军 2026-07-25 23:52 北京
据英国《经济学人》杂志23日报道，美国企业家马斯克预测，未来5年内，人工智能（AI）的整体智能水平可能超过全人类。他承认，“杀手机器人”威胁人类的风险仍然存在，但不应因此为AI技术踩下刹车。
马斯克接受英国《经济学人》杂志采访
在接受《经济学人》杂志总编辑赞妮·明顿·贝德斯采访时，马斯克表示：“5年左右是我的猜测……届时，除了做人之外，AI或许几乎在所有方面都会比人类做得更好。”他认为，最可能出现的情况是，人类将迎来一个“令人惊叹的富足时代”。
谈及AI技术是否可能失控时，马斯克表示，这种情况很可能发生。他以黑猩猩为例解释称：如果AI与人类的智力差距远大于人类与黑猩猩之间的差距，“很难想象黑猩猩能够掌控局面”。
关于AI技术的风险，贝德斯提及马斯克曾表示，“杀手机器人消灭人类的概率约为10%至20%”。对此，马斯克回应称：“我仍然认为AI和机器人存在风险。”不过，他的“哲学结论”是“往好的方面看”，“我看不到有任何办法真正阻止AI和机器人这股惊人的发展势头……即使有停止按钮，我们也不应该按下它。”
贝德斯随后以火箭发射作类比，询问如果马斯克旗下公司研发的火箭存在10%至20%的爆炸风险，他是否仍会乘坐。马斯克给出了肯定回答。他表示，既然无法完全阻止，“我们唯一能做的就是尽量降低出现糟糕结果的概率”。
马斯克还主张加强行业合作，建议领先AI企业定期交流安全问题，并在新模型发布前允许竞争对手有一至两周的审查时间。“竞争对手可以相互监督。”他说。 审核 | 卢长银 编辑 | 徐璐明 校对 | 向歆悦
请 长按下方二维码关注我们or回到文章顶部，点击环球时报 （微信公众号ID：hqsbwx）
Key Quotes
> I don't see any way to truly stop the astonishing development momentum of AI and robotics...... even if there is a stop button, we should not press it.
> About five years is my guess...... at that point, aside from being human, AI might outperform humans in nearly every aspect.
> Since we cannot completely stop it, the only thing we can do is try to reduce the probability of bad outcomes as much as possible.
> Competitors can monitor each other.
Tags
Artificial Intelligence
AI Risk
Musk
AI Prediction
AI Development
Related Articles
* Before NATO summit, Russia and Ukraine launch large-scale attacks on each other * Over 200 Old Photos Expose Unit 731 Crimes! Perpetrators Disguised as 'Kind Fathers' * 145. Oral History of SpaceX Development: Former Executive Lewis Hong on Musk's Hiring Philosophy, the Biggest IPO, Space & AI, and Prelude to Human Civilization Expansion? * AI and Careers Over the Last Three Years: Turbulence, Patterns, and Professional Portraits * Zhu Songchun: China's AI Industry, Don't Be Led Astray by 'Musk Faith' * GPT-5.6 Three‑Tier Models Fully Encircle Claude Fable 5, “Super App” ChatGPT Work Takes the Lead, which comprehensively surpass Claude Fable 5 in programming, Agent task execution, cost efficiency, and other aspects, and have disclosed detailed pricing and safety assessments.") * The Semi‑Annual Report Is Both a “Report Card” and a “Mobilization Order” * The U.S. pointed at China and then slapped itself hard... * Building an Agent from 0 to 1: Agent Principles Analysis and Personal Assistant Practice (In-Depth Long Read) * "Naked Performance" Has Gone Too Far