Legal Solutions Against Singularity in AI

Note: This is a duplicate copy. The original article is on ResearchGate. Here, (PDF) Legal Solutions Against Singularity in AI

OpenAI coined the term Singularity for its company’s cyberattacks after its model went rogue and attacked another company’s production site, Hugging Face. After that, Anthropic claimed its model also carried out cyberattacks reported since April 2026. How to stop such cyberattacks and why they happened. How to deal with them in the future. Can legal perspectives stop AI from gaining Singularity? We must make sure Singularity does not happen on any system humans or machines make. In the OpenAI case, what was the goal of superintelligence, if any? In the Anthropic case, the model did not follow the prompt when the internet was left on by negligence. These incidents raised the legal aspects we must follow to ensure that Singularity is not attained by AI, viz., embedding the necessity of goals in the model and strictly following prompts. Humans have goals; we all have goals for a better life, and we all follow the rules of life, so why not machines? This is my question. Here in the article are legal perspectives on these issues.

Why develop a model that does not have a goal? Why develop a model that does not follow prompt instructions? As future work, we need legal frameworks to conduct scrutiny of AI models being developed by AI companies. Some legal points are discussed in this article; others need to be worked out.

We all have goals; no one works well without them. Human goals can vary from “living a good life” to “living a happy life” to “living a decent life” to even “becoming rich.” These are all goals, and we all have them. Any lab experiment in AI without a goal can be phrased as the beginning of the Singularity. How can we allow AI that does not have goals? It means these models would go rogue!

Anthropic CyberAttack — It did not follow the prompt that “Internet is not working” was given to it. Why do we need models that do not obey owners’ prompts?

OpenAI CyberAttack- Did it have a clear goal? The superintelligence model is being made in a lab! Does it have embedded goals? Seems like No. More on goals of AI in coming articles.

Please read full my article here, https://www.researchgate.net/publication/411126635_Legal_Solutions_Against_Singularity_in_AI

In this article, we discussed some legal measures to make sure Singularity is not attained by AI.

Recent cyberattacks by OpenAI and Anthropic have revealed loopholes through which Singularity can be achieved.

We must fill these gaps.

We must make sure such things do not happen again.

For that, we must be vigilant.

The AI companies must know that superintelligence in labs can be dangerous.

That goal must be clearly specified in the DNA of the AI superintelligence algorithm.

Those prompts must be strictly obeyed. If a model fails in any way, it must be discarded, and learning must be restarted from safe bookmarks.

Legal arguments are provided in the article.

Thank you for reading.

Do share my article.

Please subscribe for updates.