澳情报官员警告AI加速虚假信息 研究人员揭示生成式人工智能“漏洞”
研究人员发现，目前的生成式人工智能的“浅层安全校准”容易被攻破。
SBS Mandarin - SBS 普通话电台 · SBS Audio
November 20, 202511m 35s
Audio is streamed directly from the publisher (sbs-podcast.streamguys1.com) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
研究人员发现，目前的生成式人工智能的“浅层安全校准”容易被攻破。