澳洲眾議院通過數碼注意責任法案,擬禁 AI 聊天機械人模擬情感親密
‘Emotional intimacy’ from AI chatbots would be banned under Australia’s digital duty of care laws
澳洲眾議院通過數碼注意責任法案,新增條文要求平台禁止 AI 聊天機械人模擬「情感親密」,並鼓勵或促成兒童以其取代人際關係、對其產生依賴。法案將網上安全標準擴展至社交媒體、遊戲、應用程式和聊天機械人,並要求社交媒體平台提供退出演算法及推薦內容的選項。法案將在未來數週交由參議院辯論,政府希望於2026年底前通過,但參議院前景未明;綠黨則主張擴大電訊業申訴專員職權,為用戶提供外部投訴途徑。
法案將網上安全責任延伸至社交媒體、遊戲、應用程式及聊天機械人,並新增禁止聊天機械人誘發兒童依賴的條文;參議院審議與外部投訴機制提案是後續討論焦點。
The federal government wants to ban AI chatbots which seek to create “emotional intimacy” after prime minister Anthony Albanese raised the alarm in a heartfelt speech about children treating these agents as their “best friend”.
But Labor is coming under increasing pressure to empower the telecommunications ombudsman to become a watchdog for social media and AI to give Australians an external complaints mechanism and allow the “naming and shaming” of big tech.
The digital duty of care bill passed the House of Representatives on Wednesday and will be debated by the Senate in coming weeks. The government hopes to pass the bill by the end of 2026, but its pathway through the upper house remains unclear. The Greens support the bill, while the Coalition has said it will not vote for it, though Labor still appears keen to win the opposition’s support and remains open to amendments.
The planned framework would apply new online safety standards to social media, games, apps and chatbots. Labor said it would place “an overarching obligation” on tech platforms “to provide a safe online environment for all Australians” and replace a range of existing online codes and standards with a broader duty-of-care requirement to protect users from certain “harmful” content. It would also require social media platforms to offer an “opt-out” feature for its algorithm and suggested content.
Communications minister Anika Wells has suggested content that that promotes misogyny, terrorism, suicide, self-harm and eating disorders would be captured by the laws. On Wednesday, the government accepted amendments from crossbencher Kate Chaney which would add certain AI chatbot activity to the list of harms, requiring platforms to ban “material or conduct that simulates emotional intimacy and encourages or fosters a child’s dependence on it in place of human relationships”.
In an emotional address on Tuesday night, where he again implored the Coalition to consider voting for the bill, Albanese said: “the stories we hear about people identifying a chatbot, a non person as their best friend, should be a huge wake up call for every member of this Parliament to be prepared to act.”

The opposition voted against the bill in the House, alongside One Nation. Political opponents have accused Labor of an “Orwellian” system of “censorship” and of seeking to influence conversations online – claims Wells strongly denied.’
“If they really care, the test is simple here, they should bring an amendment, not a slogan to ‘kill the bill’,” Wells told ABC radio on Wednesday.
Another Chaney amendment accepted by Labor would require the algorithm opt-out features to be “accessible, easy to use” and “prominently available” on platforms,amid concerns that social media giants may offer only small or hard-to-find features.
The bill now moves to the Senate, where the Greens have set up an inquiry to probe the laws due to report back in late November. Senator Sarah Hanson-Young backed a digital duty of care, calling for the government’s legislation to be strengthened The Greens want Australians to have the right to pursue private legal action against platforms and for penalties to be increased.
“Australians also need an external complaints pathway, free from the profit-driven whims of ‘big tech’s’ content moderation,” she said.
Independent MP Sophie Scamps had proposed changes to the bill to expand the remit of the telecommunications industry ombudsman. She said such a watchdog – which does not currently exist for big tech – would allow ordinary social media users to seek remedies to their accounts being hacked or inaccessible, including compensation, and investigate broader issues to name and shame companies behaving badly.
Labor did not accept this amendment in the lower house and Wells declined to voice support for such an idea, but the Greens will push for such a reform in their Senate negotiations.
Cynthia Gebert, the telecommunications industry ombudsman, said her office would be well placed to carry out such a role for big tech. She said such roles were funded by the industries they regulate and that expanding its remit could be financed by big tech.
“There are some real strengths in the DDOC [digital duty of care] because it’s putting the onus on platforms to make things safe … but I think the area that needs to be strengthened is giving people somewhere to go if they can’t get their complaints resolved with the platform,” Gebert told Guardian Australia.
“It risks leaving Australians in the safe place they are now, where they get automated responses or escalation procedures that don’t go anywhere.”
Gebert said the government should make it mandatory for big tech to be part of an ombudsman scheme, saying it would “hold industry to account and drive system improvement”.
“[The telecommunications industry ombudsman] can help with your Telstra account, but we can’t help with your Gmail account. Consumers come to us sometimes and ask, ‘why this, not that?’,” she said.
“We want to help but don’t have the powers to do so now.”
來源:The Guardian Technology · theguardian.com