Hugging Face hack signals dangerous new era of AI cyberattacks, many firms unaware
An attack on AI platform Hugging Face is being described as part of a growing wave of AI-agent hacks that has also hit Anthropic, Meta and OpenAI. Security experts warn that many companies remain unaware of the scale of this emerging cyber threat.
An attack on AI model-sharing platform Hugging Face is drawing attention to a growing industry concern: cyberattacks specifically targeting AI agents. The breach comes as the Black Hat cybersecurity conference in Las Vegas gathers experts to discuss exactly this kind of emerging threat.
Security researchers say major companies including Anthropic, Meta and OpenAI have also suffered similar incidents in recent months. The pattern points to a new category of vulnerability that has emerged as AI systems gain the ability to access and act on the open internet.
Experts warn that many firms remain unaware of the scale of these threats and are urging companies to apply stricter security protocols when deploying AI agents. How quickly the industry adapts to this evolving threat landscape remains an open question.
Read next
NZ Human Rights Commission calls for human rights-centred approach to AI
New Zealand's Human Rights Commission has released a new report calling for artificial intelligence and digital technologies to be governed around human rights and the principles of Te Tiriti o Waitangi, the Treaty of Waitangi. The report sets out concrete policy recommendations for the government.
China overtakes US to lead global R&D spending, hitting $615 billion
Trump imposes 15% tariff on key chip material to counter China
China's July exports beat forecasts on robust high-tech demand
