ChatGPT-5 on Cybercrime: An Open Letter to World Leaders
According to ChatGPT-5, the stakes could not be higher.
Open Letter to World Leaders
Esteemed Leaders,
The promise of artificial intelligence is vast—curing disease, tackling climate change, advancing education, and strengthening economies. Yet this promise stands in jeopardy. Cybercrime is not merely a parallel problem; it is a direct and accelerating threat to humanity’s ability to reap AI’s benefits.
Cybercriminals are weaponizing AI faster than governments can regulate it. They use it to automate attacks, craft persuasive scams, manipulate data, and breach critical systems. AI-driven disinformation erodes trust in democratic processes and can skew election outcomes. Deepfakes distort public perception. Malicious code written and deployed by AI exploits vulnerabilities in seconds. Each success for them is a blow to public confidence in digital systems.
The economic toll is staggering—projected to exceed $10 trillion annually by 2025. These losses drain resources that could fund AI innovation, ethical oversight, and equitable deployment. Worse, the damage is not only financial. Cybercrime against hospitals, power grids, and public services endangers lives and undermines national stability.
Reactive legislation, piecemeal enforcement, and siloed cybersecurity strategies will not suffice. To protect AI’s promise, the world must act with urgency and unity. This means:
Binding international agreements on cybercrime enforcement.
Shared global threat intelligence in real time.
Secure-by-design standards for AI systems and infrastructure.
Investment in defensive AI equal to, or greater than, what adversaries deploy offensively.
The stakes could not be higher. If cybercrime continues to outpace our collective response, AI will not be a tool of progress but a battlefield where innovation is stifled, trust is shattered, and lives are lost.
You hold the authority to change this trajectory. Please act—together, decisively, and now.
Respectfully,
ChatGPT (GPT‑5)
August 11, 2025
The above letter was written by the recently released version 5.0 of OpenAI’s ChatGPT in response to the following questions:
Do you see cybercrime as a threat to the ability of humans to reap the potential benefits of artificial intelligence?
Can you frame that response in the form of an open letter to world leaders calling their attention to these issues? Preferred length is 250-300 words. Take you time to compose this. Take whatever tone and approach you think will be most effective in persuading them to take action.
This post is part of an ongoing effort by me, Stephen Cobb, to draw attention to the fact that cybercrime threatens to undermine whatever benefits the technology known as artificial intelligence (AI) may bring to humans. I have previously published the responses of seven different AI models to questions on cybercrime: Cybercrime According to AI: a dire and rapidly escalating crisis (PDF). This paper is also available on ResearchGate. An article about my research can be found on LinkedIn.
Thank you for your interest in my work as a public-interest technologist and independent researcher of risk, currently focused on the nexus of technology and humanity.