玩命加载中...
玩命加载中...
This video features a conversation with Rachel Tobac, an ethical social engineer who demonstrates how human vulnerabilities are often easier to exploit than technical systems. She begins by revealing the host's personal details—full name, hometown, and past hobbies—gathered from open‑source intelligence (OSINT), shocking him despite his cautiousness. Rachel shares her background: a Defcon social engineering contest participant with no security degree, she leveraged improv skills to build rapport and extract sensitive information, founding Social Proof Security after early successes. The discussion then shifts to common organizational weaknesses, such as poor identity verification, MFA fatigue attacks, and the dangers of password reuse. Recommendations include moving from knowledge‑based authentication to app‑based or hardware tokens like YubiKey.
这段视频与道德社会工程师 Rachel Tobac 进行对话,她展示了人类漏洞往往比技术系统更容易被利用。她首先通过开源情报(OSINT)收集并透露了主持人的个人详细信息——全名、家乡和过去的爱好,令尽管谨慎的主持人震惊。Rachel 分享了她的背景:作为 Defcon 社会工程竞赛的参与者,她没有安全学位,凭借即兴表演技巧建立融洽关系并提取敏感信息,在早期成功后创立了 Social Proof Security。讨论随后转向常见的组织弱点,如身份验证不严、MFA 疲劳攻击以及密码复用的危险。建议包括从基于知识的认证转向基于应用或硬件令牌(如 YubiKey)的认证。
The hosts explore OSINT challenges, noting that even after extensive data removal, AI‑powered reverse image searches can uncover decades‑old photos from local newspapers or relatives' social media. A live demonstration showcases voice cloning: using a short audio sample, Rachel clones the host's voice, spoofs his caller ID, and tricks a friend into revealing a security question answer in real time. The conversation highlights how quickly attackers can now impersonate trusted contacts with minimal voice data, urging verification through a secondary channel.
两位主持人探讨了 OSINT 的挑战,指出即使进行了大量数据删除,AI 驱动的反向图像搜索仍能从当地报纸或亲属的社交媒体中找出几十年前的照片。现场演示展示了语音克隆:使用一段简短的音频样本,Rachel 克隆了主持人的声音,伪造了他的来电显示,并实时骗过一位朋友,使其透露了一个安全问题的答案。对话强调了攻击者现在能以极少的声音数据快速冒充可信联系人,敦促通过辅助渠道进行验证。
Broader societal risks are examined, including AI chatbots reinforcing delusions ('AI psychosis') and the dangers of always‑on AI companions like the 'Friend' pendant, which may erode social skills and enable sycophantic behavior. Rachel criticizes placing the cybersecurity burden on users and calls for corporate responsibility, advocating for agentic AI to handle content moderation—a job she once nearly had at Facebook where exposure to horrific imagery was routine. She concludes by advising people to be 'politely paranoid,' cross‑verifying suspicious requests because scammers rely on victims not double‑checking through another method.
还审视了更广泛的社会风险,包括 AI 聊天机器人强化妄想(“AI 精神病”),以及像“Friend”吊坠这类始终在线的 AI 伴侣的危险,它们可能侵蚀社交技能并助长奉承行为。Rachel 批评将网络安全负担推给用户,呼吁企业承担责任,主张使用代理 AI 处理内容审核——她曾在 Facebook 差点获得这样一份工作,那里接触恐怖图像是家常便饭。她最后建议人们保持“礼貌的偏执”,对可疑请求进行交叉验证,因为骗子正是利用受害者不会通过其他方式再次确认。
我相信每个人都会被黑客入侵除了你。你已启用了两步验证。你所有的密码都非常强大。你从不点击可疑链接。你做所有事情都对。但你仍然不能免受这个人的攻击。>> 好的。所以,在我最初几次入侵一个组织时,他们想让我给他们的一位高管打电话,试图获取信息。所以,如果你想入侵一位高管,你实际上必须联系他们的行政助理。所以我查出了谁是他们的EA。我给他们打了电话,我在30秒内就拿到了信息。我需要的那些信息基本上是用来从公司偷钱。>> 那就是Rachel Tobach。她是一名社会工程师,这意味着她不是入侵电脑,而是入侵人,而且她一定是有史以来最优秀的之一。我有机会坐下来和她聊聊她如何入侵大公司、AI精神病,我还给了她一个小小的挑战。我想让她对我做些研究,看看我是否会被入侵,因为说实话我曾是那种人,对吧?我不认为自己有风险。我不认为自己可能被入侵。但是,呃,天哪,我大错特错。>> 那么,Rachel Tobac,>> 嗯,>> 老兄,我知道我早知道会这样,但还是让我措手不及。太神奇了。好吧,如果我没说错。是的,那个。>> 再说一遍。>> 来自弗吉尼亚。>> 哦不。哦不。>> 给我讲讲Destination Imagination。>> 你能告诉我关于那些魔术你过去会做的东西,或者那些催眠表演,你在舞台上做的?>> 这太让我觉得不可思议了。>> 你甚至不愿意看着我的脸。>> 我不能,因为它是如此。听着,在这个空间里,没有人叫过那个名字,大概三年了。>> 是的。>> 所以,听到它感觉很奇怪。它已经到了我甚至不觉得那是我的名字有时候。>> 我想如果我像那样称呼你用那个名字,你也不会回应。>> 是的。
AnyDesk is incredible and one of the biggest contributors in helping fight back against scammers! To learn more about AnyDesk, visit https://anydesk.com/en Follow Rachel here: X: https://www.twitter.com/racheltobac Instagram: https://www.instagram.com/racheltobac LinkedIn: https://www.LinkedIn.com/in/racheltobac https://www.socialproofsecurity.com Help us fight these scammer! Become a member today: https://www.youtube.com/channel/UCC9EjyMN_hx5NdctLBx5X7w/join #scammers #scambait #scam All of our Socials! Become a member: http://bit.ly/3uRvoEy Official Site: https://bit.ly/3RgsGoG Merch: https://bit.ly/4cT6Ql8 Facebook: https://www.facebook.com/PierogiSP Twitch: http://bit.ly/3bgVGrX Twitter: https://bit.ly/3c00Jw5 Instagram: https://bit.ly/3c0ZnkS Scammers typically target the elderly and attempt to steal money by several means. 1. Your Bank Saving or Checking accounts 2. Investment accounts or 401k retirement funds 3. Credit and Debit cards 4. Purchasing Gift cards 5. Cash withdrawls 6. Cryptocurrency Scammers are ruthless criminals that have no problem stealing every last dime from their victims bank accounts. Please protect yourself as well as family members who might not be familiar with these type of scams.