{"product_id":"human-compatible","title":"Human Compatible","description":"\u003col\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eMany believe \u003c\/span\u003e\u003cspan\u003econflict between humans and machines is inevitable\u003c\/span\u003e\u003cspan\u003e, with a predictable outcome.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eStuart Russell\u003c\/span\u003e\u003cspan\u003e, a distinguished AI researcher, argues that this scenario can be avoided, but only with a fundamental rethink of AI.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eRussell explores the concept of \u003c\/span\u003e\u003cspan\u003eintelligence in both humans and machines\u003c\/span\u003e\u003cspan\u003e, laying the groundwork for his argument.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eAI has significant \u003c\/span\u003e\u003cspan\u003enear-term benefits\u003c\/span\u003e\u003cspan\u003e, including intelligent personal assistants and advancements in scientific research.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eHowever, \u003c\/span\u003e\u003cspan\u003eAI breakthroughs are still required\u003c\/span\u003e\u003cspan\u003e before superhuman AI becomes a reality.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eRussell discusses how humans are already finding ways to \u003c\/span\u003e\u003cspan\u003emisuse AI\u003c\/span\u003e\u003cspan\u003e, such as through lethal autonomous weapons or viral sabotage.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eIf superhuman AI emerges, it would possess \u003c\/span\u003e\u003cspan\u003efar greater power than humans\u003c\/span\u003e\u003cspan\u003e, raising concerns about control.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eRussell suggests \u003c\/span\u003e\u003cspan\u003erebuilding AI from a new foundation\u003c\/span\u003e\u003cspan\u003e where machines are uncertain about human preferences, preventing them from overriding our goals.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cdiv\u003e\n\u003cspan\u003eThis new approach would create \u003c\/span\u003e\u003cspan\u003ehumble, altruistic machines\u003c\/span\u003e\u003cspan\u003e that are committed to pursuing human objectives, ensuring they remain safe and beneficial.\u003c\/span\u003e\n\u003c\/div\u003e\n\u003c\/li\u003e\n\u003c\/ol\u003e","brand":"The Book Shop Nepal","offers":[{"title":"Default Title","offer_id":47964940271871,"sku":null,"price":560.0,"currency_code":"NPR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0735\/5452\/6463\/files\/download-2026-01-01T144450.323.jpg?v=1767258019","url":"https:\/\/bookshopnepal.com\/products\/human-compatible","provider":"The Book Shop Nepal","version":"1.0","type":"link"}