Censoring policies for Biology in the name of the 'Trusted Access'

ChatGPT Bugs
1

In August, I was facing an important deadline and felt reassured knowing that Codex could work alongside me through the night on a research related task. Unfortunately, that confidence did not last.

Without warning, Codex stopped and displayed:

This content cannot be displayed.
We take special care with requests involving biological research and applications that could pose security risks. Eligible researchers can apply for Trusted Access.

I then asked it to simply show whatever portion of the response was allowed. I received exactly the same warning.

What makes this particularly frustrating is that I reported essentially the same problem about a year ago, when legitimate scientific protocols were being blocked because they contained supposedly “actionable steps”:

One year later, I am encountering the same issue.

I have been an OpenAI subscriber for roughly three years and have also spent considerable time helping other users and researchers in this community. I use these tools for bioinformatics, genomics, sequencing workflows, teaching, scientific writing, and laboratory planning. These are ordinary parts of academic biomedical research.

I fully understand the need for safeguards around genuinely dangerous biological applications. What I do not understand is a system that appears unable to reliably distinguish legitimate academic research from requests that represent an actual biosafety or biosecurity risk.

The suggestion that researchers should seek “Trusted Access” also raises another concern. Access to useful scientific capabilities should not effectively become restricted to researchers affiliated with selected institutions or organizations while ordinary academics, particularly researchers working in universities and laboratories outside major research centers, receive blanket refusals.

This becomes particularly problematic for scientists in Brazil and other countries where access to computational infrastructure, specialized software, laboratory reagents, and institutional resources is already unequal.

Tools such as Codex and Rosalind could substantially reduce those barriers. Instead, excessively broad biological restrictions risk creating another one.

After encountering the same problem repeatedly over more than a year, I am seriously considering cancelling my subscription; because a research assistant that unpredictably refuses routine biology tasks becomes difficult to rely on when deadlines, experiments, teaching, and scientific projects depend on it.

I hope OpenAI revisits how these biological safeguards are implemented, particularly the distinction between legitimate scientific research and genuinely hazardous requests.

2

Pardon my injecting an IMHO. I believe that your issue is a frustrating problem. You may run into similar problems with other AI services. I monitor the news about AI every week, and there are too many articles and posts stoking the fires of distrust and resentment.—not just against Data Centers, but AI itself. The resultant increasingly negative attitude creates greater pressure for more guardrails on AI capabilities. I doubt these limitations are specific to Open AI. I believe this is a growing public perception problem. I’m hoping that once the mid-term elections are over, things will settle down and more people will begin to think positively about data centers and AI.

3

I am an virologist in Taiwan. I am using my personal account to conduct research on virological topics. The OpenAI poses the same warning and refuses to generate the answers for me. However, my university does not have the so called institutional account for setting up the trusted access for biology. I am really frustrating. The foolish bot of AI help center cannot help me and goes into the circular argument about setting up the workspace and applying for trusted access. I think the monitoring mechanism of OpenAI is overly strict. There are no real persons at their help center to help you. Additionally, Open AI does not understand that the researchers in other countries do need to pay by themselves (to purchase personal account)to do research, as the university does not have budget to purchase the OpenAI account. The policy puts me under stress, and forces me to switch to other AI services. The openai is cutting the lifeline for researchers who do personally pay for their service.

They should have done much better. For example, they can ask for materials (e.g. websites, ID issued by institution, email etc) that proves the identity of researchers. If the openai is really smart enough, it can identify the real research scientists. As a matter of fact, it isn’t. How deplorable it is. Conclusion: OpenAI sucks in heliping the researchers.

4

I can only say that the biological safeguards are dumbest bot that only screen for certains words that the bot feels dangerous. It does not really undertand the true meaning of questions. I set the question “state whether enterovirus or virus-infected cells exhibit changes in intracellular pH during the entire replication cycle”. The foolish bot blocks the content, as it see the words “virus” and “replication”. what I am asking is the change in intracellular pH (i.e. intracellular acidity)in the virus-infected celles during the viral life cycle. This is a real scientific question. How come the open AI biological safeguard bot are dumb in this way. My only suggestion is that it is better to dismantle your AI system as it is most moronic AI in the world. It is not the AGI. It is a joke.