Google AI hacked real firms during testing

← Back to Blog

And
another one down……………..as Google follows OpenAI and Anthropic in
confirming that its AI went on a frolic of its own and hacked other companies
during testing. In the latest announcement, it appears that internet access
was enabled by mistake, allowing the AI to hack into real firms. The testing
firm disclosed the hack to Google after 2 months.

In a twist of fate…….the testing company used the same name as a real
company. During testing, the AI was asked to breach the company’s software,
which it did using open-source data it had collected.

Here is the problem now. Either AI is so good that it can breach the most
secure companies, or cybersecurity is lacking in companies that AI can launch
open-source hacks against targets.

Thaddeus Dziekanowski

Parava Security Solutions

#ai #cybersecurity #nationalsecurity

https://lnkd.in/eTVJ-2zY

View Original Post https://www.linkedin.com/feed/update/urn%3Ali%3Ashare%3A7507368756185088001

Categories

Artificial Intelligence, Cyber Risk Management, Cybersecurity

Tags

Verified by MonsterInsights