SplxAI has raised $7 million to enhance security for AI systems, recognizing the growing threats posed by generative AI which many believe gives cyber attackers an advantage. In response, the Croatian startup aims to redefine vulnerability testing by implementing proactive measures against threats. Traditional red-teaming methods, which simulate attacks on AI systems, can be time-consuming. So, to streamline this process, SplxAI relies on adjusting system prompts preemptively to mitigate risks, allowing faster and safer deployment of AI tools in businesses.
AI systems are vulnerable to all kinds of new threats, from data poisoning to adversarial attacks.
In a recent Accenture survey of 600 bank cybersecurity executives, four in five said generative AI is helping hackers faster than banks can keep up.
SplxAI takes an offensive approach by adjusting system prompts - guidelines that shape how an AI model responds to user queries.
Companies are racing to adopt AI in pursuit of productivity and profits, but the last thing anyone wants is a chatbot going off the rails.
Collection
[
|
...
]