与「AIEthics」相关的搜索结果

每周六给大家分享一篇深度好文： 本周：《There’s this deep mystery of what, actually, is this thing?: the philosopher inside Google DeepMind AI》 作者：Robert P Baird 平台：The Guardian 文章链接放在文末！ 推荐对 AI 有兴趣或者对 AI 投资下注的朋友一定要看下，是一篇很好的深度人物 + 思想随笔。 推荐的主要原因，是因为它写的是 Google DeepMind 里的哲学家 Iason Gabriel： 这个结合很有趣，一个政治哲学背景的人，如何在最前沿的 AI 实验室里，思考 AI 到底应该服务谁、服从谁、伤害谁、又由谁来决定它的价值边界。 文章从 Gabriel 2017 年加入 DeepMind 写起。 当时他几乎是少数真正进入前沿 AI 实验室内部的哲学家之一。文章借他的经历，把 AI 领域里两个长期分裂的阵营讲清楚了：一边是关注“失控、对齐、超级智能风险”的 AI safety，一边是关注“偏见、权力、社会伤害”的 AI ethics。 所以你可以看到 AI 对齐不是简单地让机器“听人话”，而是一个更复杂的四方关系：AI 系统、用户、开发者、社会。 如果 AI 只服务开发者，可能会损害用户；如果 AI 只服从用户，也可能损害社会；如果只追求技术最优，又可能忽略人的价值冲突。 另外关于 Agenty 的普及肯定是不可逆的趋势，未来 AI 不再只是输出文字，而是会在现实世界里行动。这个变化会让伦理、责任、权力边界都变得更尖锐。 与我而言这篇内容的最大启发： 1️⃣AI 时代更应该多读书多思考 AI 时代不要只追新工具，工具只是形式上的，真正重要的是你有没有自己的“价值排序”和“决策边界”。 没有这个边界，AI 越强，只会把你的混乱放大。 2️⃣超级个体； 未来的超级个体，除了熟练掌握工具，更重要的，是必须是能把 AI 放进自己的判断系统里，同时知道什么时候不能让 AI 替自己做决定的人。 3️⃣投资启发； 我以前以为 AI 竞争的除了技术就是算力。 原来 AI 公司、Crypto 项目、平台型产品，最后竞争的不只是技术参数，而是谁掌握入口、数据、执行权和价值定义权和伦理决定权。 谁能决定“系统应该偏向谁”，谁就拥有真正的权力。 AI 还有很长的发展路径，投资一定要多样化，不要只看某一个面。 文章链接：

🚨🤖 PARTNERSHIP INTERVIEW: 51% OF THE INTERNET IS BOTS, CAN HUMANITY STILL WIN? More than half of online traffic today is not human. It is bots, and a third of them are malicious. AI is generating fake profiles, fake comments, even fake people. And if Big Tech owns your identity, one hack could erase you from the internet entirely. @terencekwok, founder of our portfolio company, Humanity Protocol, is building a system to change that, giving people control of their digital identities through blockchain verification. We talk about: •⁠ ⁠What happens when bots outnumber humans online •⁠ ⁠How Apple, Google, and Facebook control who we are •⁠ ⁠Why digital ID could be the most important tech of the decade •⁠ ⁠How the Mastercard partnership could redefine online identity •⁠ ⁠And why the next big war might be between real humans and fake ones Terence Kwok says the future of the internet depends on one thing. Knowing who is real and who is not. 00:00 – Over half of all web traffic is now bots 01:00 – Why identity matters in an AI-driven world 01:45 – Terence Kwok on losing trust in digital life 03:59 – When social media can’t tell what’s real 04:20 – Big Tech controls your existence 05:26 – The Humanity Protocol solution 08:02 – The “dead internet” is already here 08:45 – Bot traffic and the $100B fraud economy 10:15 – AI-generated content is replacing humans 12:00 – AI dependence erodes human skill 13:10 – Verifying truth through credentials 14:01 – What it means to be human online 15:17 – When AI agents need identities too 17:56 – Can AI resist shutdown or blackmail 19:19 – Why every AI needs a human “owner” 21:23 – Self-driving cars as potential weapons 22:09 – Infrastructure under AI control 23:30 – Artificial muscles stronger than humans 24:26 – Teaching AI values like a child 25:40 – Training on humanity’s worst data 26:44 – Does intelligence lead to compassion 26:46 – Who’s winning the AI race: U.S. or China 28:42 – China’s real-world AI advantage 29:31 – Everyday life inside China’s AI economy 31:46 – U.S. regulation vs. China’s speed 32:21 – Different strengths, different outcomes 32:50 – Bringing AI ethics back to Humanity 33:02 – Building Humanity with blockchain partners 35:21 – Mastercard and on-chain verification 36:26 – Control and monetize your own data 37:38 – Real-world utility over hype tokens 39:50 – Decentralized identity for a human-led web 39:59 – “Digital identity is the foundation of the next internet” Disclaimer: We are investors in this project and receive benefits from our involvement. This content is for promotional purposes only and is not financial or investment advice. Cryptocurrency carries significant risk, and you should always do your own due diligence before making any decision.