Cybersecurity: AI attacks and hijacking
● AI and generative AI systems can be easily hijacked to generate malicious code, even when designed to reject such requests.
● Other types of attacks, known as "model evasion attacks," exploit modified inputs to cause unexpected behaviours in AIs, such as making a self-driving car misinterpret traffic signs.
● Poisoned data can introduce backdoors into AI models, leading to unintended behaviours, which is concerning due to the lack of control engineers have over their data sources.
Watch the video
● Other types of attacks, known as "model evasion attacks," exploit modified inputs to cause unexpected behaviours in AIs, such as making a self-driving car misinterpret traffic signs.
● Poisoned data can introduce backdoors into AI models, leading to unintended behaviours, which is concerning due to the lack of control engineers have over their data sources.
Udio, Suno: AI-generated music is already competing for low added value orders
Read the article
Ethical AI and children: the benefits of a multi-disciplinary approach
Read the article
Search and rescue: drones that detect human voices under collapsed buildings
Read the article
Neurotechnology: auditory neural networks mimic the human brain
Read the article
P-C. Langlais (PLEAIS): “Our language models are trained on open corpora”
Read the article