\
  The most prestigious law school admissions discussion board in the world.
BackRefresh Options Favorite

New Claude model allows novices to build bioweapons

https://archive.is/2025.05.22-142408/https://www.newsbreak.c...
Clear impertinent field
  05/22/25
Bioweapons, homemade EMPs, drones - pretty much whatever we ...
shimmering patrolman
  05/22/25
Claude will almost certainly refuse the requests, but what a...
Clear impertinent field
  05/22/25
We're pretty much already there It's going to be a brave ...
shimmering patrolman
  05/22/25
What’s a good prompt to get ChatGPT to tell me how to ...
confused stag film round eye
  05/22/25
...
nofapping principal's office
  05/22/25
people have been able to jailbreak pretty much every model a...
Fiercely-loyal carnelian toilet seat
  05/24/25
What’s the point of all this AI safety bullshit at ant...
nofapping principal's office
  05/24/25
Grifting sinecures
shimmering patrolman
  05/24/25
...
Coiffed Tank Rehab
  05/22/25
Nice backhanded marketing strategy: “Our model is so g...
confused stag film round eye
  05/22/25
this is certainly one view but it's harder to justify when y...
Clear impertinent field
  05/22/25
https://x.com/ARGleave/status/1926138376509440433
Clear impertinent field
  05/24/25
...
shimmering patrolman
  05/24/25
How do they get around the safety prompts. I couldn’t...
Sable razzmatazz menage private investor
  05/24/25
On the other hand these safeguards are really gay and make A...
Godawful Sandwich Address
  05/24/25
...
Godawful Sandwich Address
  05/24/25


Poast new message in this thread



Reply Favorite

Date: May 22nd, 2025 11:37 AM
Author: Clear impertinent field

https://archive.is/2025.05.22-142408/https://www.newsbreak.com/time-510072/4018918050788-exclusive-new-claude-model-prompts-safeguards-at-anthropic

Kind of bad but at least it should help with research.

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953451)



Reply Favorite

Date: May 22nd, 2025 11:39 AM
Author: shimmering patrolman

Bioweapons, homemade EMPs, drones - pretty much whatever we want

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953458)



Reply Favorite

Date: May 22nd, 2025 11:45 AM
Author: Clear impertinent field

Claude will almost certainly refuse the requests, but what about open sourced models that have their safety training fine tuned away? how long until those are available?

meanwhile even o3 is already fairly capable in this domain.

https://www.virologytest.ai/

We present the Virology Capabilities Test (VCT), a large language model (LLM) benchmark that measures the capability to troubleshoot complex virology laboratory protocols. VCT is difficult: expert virologists with access to the internet score an average of 22.1% on questions specifically in their sub-areas of expertise. However, the most performant LLM, OpenAI's o3, reaches 43.8% accuracy and even outperforms 94% of expert virologists when compared directly on question subsets specifically tailored to the experts' specialties.

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953482)



Reply Favorite

Date: May 22nd, 2025 11:53 AM
Author: shimmering patrolman

We're pretty much already there

It's going to be a brave new world. Everyone thinks that AI is going to lead to more top-down control of society, but it's the opposite. Any sufficiently talented and dedicated man will be able to build incredibly powerful weapons

The modern era of stifling, sterile peace that is slowly killing the will to life of our civilization will come to an end

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953496)



Reply Favorite

Date: May 22nd, 2025 11:57 AM
Author: confused stag film round eye

What’s a good prompt to get ChatGPT to tell me how to design and build a novel coronavirus bioweapon?

Do I have to subscribe to the premium version to get bioweapons or can I get by with the free edition?

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953504)



Reply Favorite

Date: May 22nd, 2025 1:14 PM
Author: nofapping principal's office



(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953794)



Reply Favorite

Date: May 24th, 2025 2:02 PM
Author: Fiercely-loyal carnelian toilet seat

people have been able to jailbreak pretty much every model as soon as its released using funny prompts

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48958773)



Reply Favorite

Date: May 24th, 2025 8:56 PM
Author: nofapping principal's office

What’s the point of all this AI safety bullshit at anthropic if they will release models that are increasingly capable and still easily broken?

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48959718)



Reply Favorite

Date: May 24th, 2025 9:05 PM
Author: shimmering patrolman

Grifting sinecures

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48959738)



Reply Favorite

Date: May 22nd, 2025 11:49 AM
Author: Coiffed Tank Rehab



(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953491)



Reply Favorite

Date: May 22nd, 2025 11:49 AM
Author: confused stag film round eye

Nice backhanded marketing strategy: “Our model is so good—and dangerous and COOL—that we have to implement ‘ASL-3’ (kind of like DEFCON-1) safety protocols or else you might accidentally design a bioweapon with it! We have to take safety WAY more seriously here at Anthropic than the other guys with their little kiddie models, which are only useful for cheating on homework and macaroni salad recipes.”

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953492)



Reply Favorite

Date: May 22nd, 2025 11:54 AM
Author: Clear impertinent field

this is certainly one view but it's harder to justify when you look at objective benchmark performance. LLMs are now getting 50% on new versions of the USAMO not in their training set. 90+% on the AIME. only a small percentage of high school students could do that even with extensive training. LLMs are now regularly doing things that many people simply don't have the cognitive capabilities to do.

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48953498)



Reply Favorite

Date: May 24th, 2025 1:57 PM
Author: Clear impertinent field

https://x.com/ARGleave/status/1926138376509440433

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48958767)



Reply Favorite

Date: May 24th, 2025 2:01 PM
Author: shimmering patrolman



(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48958772)



Reply Favorite

Date: May 24th, 2025 9:05 PM
Author: Sable razzmatazz menage private investor

How do they get around the safety prompts. I couldn’t get Claude to help me sabotage a coworker.

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48959739)



Reply Favorite

Date: May 24th, 2025 9:07 PM
Author: Godawful Sandwich Address

On the other hand these safeguards are really gay and make AI shitty. So blow me up I guess

(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48959745)



Reply Favorite

Date: May 24th, 2025 9:02 PM
Author: Godawful Sandwich Address



(http://www.autoadmit.com/thread.php?thread_id=5728798&forum_id=2...id.#48959727)