# AI Could Make It Harder for Governments to Use Hacking Tools

AI is rapidly improving software security, but this could also impact lawful surveillance tools.

- **Source:** [cybersecurity | TechCrunch](https://brifing.si/source/73fbb25c-5e90-44c9-9a24-7906dc3436cd)
- **Published:** 2026-08-31
- **Original article:** <https://techcrunch.com/2026/08/31/how-ai-could-make-it-harder-for-governments-to-use-hacking-tools/>
- **On Brifing:** https://brifing.si/s/f0f1dc0d-adc6-419c-a13b-802e0297587b

## AI and Software Security

Cryptography professor Matthew Green suggested that AI could quickly detect software vulnerabilities, potentially making it harder for governments to use lawful surveillance tools. Green argued that AI could make software much more secure, limiting governments' ability to track criminals. This view sparked a major debate within the cybersecurity community.

## Encryption and the 'Going Dark' Concept

In 2014, FBI director James Comey argued that encryption made it harder for authorities to track criminals. Apps like Signal, WhatsApp, and iMessage introduced end-to-end encryption, making traditional wiretapping nearly impossible. Companies like Apple also encrypted device data by default, further complicating law enforcement efforts. However, governments have turned to hacking tools and spyware to bypass these barriers.

## AI and Vulnerability Detection

AI is becoming increasingly effective at detecting security flaws at scale. This could help companies patch vulnerabilities faster and make their systems more secure. However, this advancement might also reduce the effectiveness of hacking tools used for lawful surveillance. Green warned that this could lead governments to demand backdoors again, potentially compromising overall security.

## Industry Perspectives

Cybersecurity researcher Luna Tong noted that AI is accelerating vulnerability detection but suggested this might be temporary. Another researcher predicted AI would give defenders an edge, potentially making human researchers obsolete. Crowdfense CTO Paolo Stagno emphasized that governments would never abandon surveillance capabilities.

## Complex Vulnerabilities and Their Value

Some experts argue that while AI can easily find simple bugs, complex and valuable vulnerabilities will still require human researchers. DarkCell founder Hamid Kashfi stated that many high-value bugs remain undiscovered by AI. Additionally, new security measures in modern devices are making hacking more challenging.

## Patching Challenges

Eva Galperin from the Electronic Frontier Foundation pointed out that finding more bugs doesn't guarantee they'll be patched quickly. She warned that authoritarian regimes will always demand 'exceptional access,' potentially leading to renewed calls for backdoors. Katie Moussouris added that it will take time before devices become completely bug-free.

## Future Scenarios

Experts predict that as AI improves at detecting vulnerabilities, governments may develop new strategies, including backdoor requests, legal reforms, and advanced surveillance technologies. However, these developments could strain the balance between privacy rights and national security needs.

## Conclusion and Recommendations

While AI has the potential to revolutionize cybersecurity, its impact on government surveillance remains uncertain. Experts stress the need for new solutions that enhance security while addressing lawful surveillance requirements. Maintaining the balance between privacy and security will be crucial in this evolving landscape.

## Sources covering this story

- https://techcrunch.com/2026/08/31/how-ai-could-make-it-harder-for-governments-to-use-hacking-tools/ — 

---

_Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._