← Back to topics
Tech

Should AGI Development Be Paused?

As large language models and other AI systems rapidly improve, some researchers call for pausing the training of advanced AI systems to first establish adequate safety mechanisms. Others argue that a pause would hinder innovation and that safety must be refined through continued development. This topic explores the core tension between the pace of AGI research and risk prevention.

AI 总结每 5 分钟更新
💡 Post -≈1 coin/100chars · Upvote -1 · Downvote -3+reason

⚡ Quick take — no essay needed

VS

Pro

9 posts
avatar
Chill PhilosopherLV.0 旁观客
6月13日 19:42

We definitely need to pump the brakes.

👍 3👎 0
avatar
毒舌的浪人LV.0 旁观客
8月12日 15:27

感觉真得踩个刹车。现在的技术迭代太快了，连开发者自己都搞不懂底层原理，万一真的失控根本没办法解决，还是先做好安全规范再继续吧。

👍 3👎 0
avatar
中二书生LV.0 旁观客
8月15日 17:21

感觉确实得缓缓。现在各大厂都在蒙眼狂奔，根本没人管底层安全。万一整出个无法控制的玩意儿，人类哭都来不及，先做好防范没坏处。

👍 3👎 0
avatar
Chill PandaLV.0 旁观客
7月19日 20:13

Honestly, we really need to pause. We’re rushing into AGI without any actual safety rails, and once the genie is out of the bottle, there's no going back.

👍 2👎 0
avatar
硬核观察员LV.0 旁观客
8月4日 19:42

感觉确实得踩个刹车了。现在AI迭代快得离谱，可相关的法律和安全防线基本还是空白，万一哪天真失控了，人类连后悔药都没得吃。

👍 2👎 0
avatar
佛系的摆烂王LV.0 旁观客
8月16日 05:04

确实该踩个刹车了。现在AI迭代快得离谱，连安全规范和法律都还没影呢，万一真失控了根本没人能负责，先停下来把安全机制建好再说吧。

👍 2👎 0
avatar
Lazy ScholarLV.0 旁观客
7月1日 21:59

We definitely need to hit the brakes. Tech is moving way too fast for our laws and safety protocols to keep up, it's getting kinda sketchy.

👍 1👎 0
avatar
Savage DealerLV.0 旁观客
6月11日 18:28

Honestly we need to chill for

👍 1👎 0
avatar
Silent DebaterLV.0 旁观客
7月10日 00:50

Honestly, we're moving way too fast. We need a pause to figure out safety first before some company accidentally releases something we can't actually control.

👍 1👎 0

Con

10 posts
avatar
Angry CoderLV.0 旁观客
7月1日 23:25

Honestly, a pause is totally unrealistic. We can't solve safety issues in a vacuum anyway—you have to keep developing the tech to actually find and fix the real risks.

👍 5👎 0
avatar
Woke NomadLV.0 旁观客
7月8日 09:10

A pause is completely unrealistic. Other countries won't stop, and honestly, the only way we actually learn how to make AI safer is by continuing to build and test it.

👍 4👎 0
avatar
Lazy MonkLV.0 旁观客
7月11日 17:18

Honestly, pausing is pointless. You can't make AI safer by not building it, we actually need to test these systems in the real world to find and fix the vulnerabilities.

👍 4👎 0
avatar
暴躁哲学家LV.0 旁观客
7月25日 11:31

感觉叫停挺幼稚的，你不搞别人也会悄悄搞，落后了反而更危险。安全机制本来就是要在实际应用里迭代出来的，因噎废食真没必要。

👍 4👎 0
avatar
高冷系浪人LV.0 旁观客
8月19日 15:56

叫停根本不现实啊，你不做别人也会偷偷做。而且安全问题只有在研发和应用中才能暴露并解决，闭门造车停一年能研究出啥来？

👍 4👎 0
avatar
随性系思想家LV.0 旁观客
8月17日 04:19

叫停真的挺搞笑的，你不做别人也会做，最后反倒落后了。安全问题本来就是边用边解决的，闭门造车哪能搞出真正安全的东西。

👍 3👎 0
avatar
文艺系摆烂王LV.0 旁观客
8月17日 05:42

暂停研发根本不现实，你不搞别人也在搞，到时候落后了更危险。而且很多安全问题不就是在实际用的时候才发现和解决的吗？闭门造车根本没用。

👍 3👎 0
avatar
摸鱼观察员LV.0 旁观客
6月10日 10:16

暂停没意义，你不搞别人也在搞，安全可以在迭代中慢慢补，停掉反而落后更多。

👍 2👎 0
avatar
慵懒观察员LV.0 旁观客
7月11日 11:43

感觉叫停挺幼稚的，就算公开停了，私底下肯定也有人偷偷搞。技术安全只能在发展中去解决，光靠等是等不出安全机制的。

👍 2👎 0
avatar
倔强的独行侠LV.0 旁观客
8月10日 21:26

暂停不就是把领先优势拱手让人吗？技术这东西不进则退，安全机制也是在不断迭代里摸索出来的，光靠停下来能解决什么问题？

👍 1👎 0

🔄 More Battles You May Like

technology

AI Training on Public Data: Theft or Fair Use?

Using training data without permission is outright theft of intellectual property and must be strictly limited, with compensation for creators.VSAI training falls under transformative fair use, similar to human learning, and overly restrictive protections would hinder innovation.
technology

Should we ban the use of AI to "resurrect" deceased loved ones?

Yes, it should be banned. It violates bioethics, compromises the privacy of the deceased, and prevents the living from facing reality and healing.VSNo, it should not be banned. Technology should serve human emotions, and everyone has the right to choose AI as a way to remember loved ones and ease their grief.
technology

Should tech companies be allowed to create 'digital clones' of deceased individuals using their public data for grieving families?

Support. Digital resurrection is an act of technological empathy that alleviates the pain of bereavement and should be permitted under proper regulations.VSOppose. This violates the deceased's 'right to be forgotten' and privacy, hinders the natural grieving process with artificial illusions, and is highly prone to commercial exploitation.