At this week’s Virus Bulletin 2024 convention in Dublin, Eire, Sophos principal knowledge scientist Younghoo Lee will current a paper on SophosAI’s analysis into making use of giant language mannequin (LLM) synthetic intelligence to the advanced downside of e mail and internet content material classification. Lee’s presentation will element the usage of “multimodal” AI—utilizing fashions to look at each the textual content and pictures in e mail and internet content material to categorise the content material as benign, undesirable or dangerous.
In his paper, “Multimodal AI: The Sixth Sense for Cyber Defence,” Lee explains that adversaries can use generative AI to create convincing phishing emails and web sites, as demonstrated in SophosAI’s “Scampaign” analysis revealed final 12 months. Moreover, conventional machine studying approaches to phishing and spam detection constructed with coaching units of present malicious and undesirable (“spam”) content material may miss new strategies that don’t match present patterns.
Nevertheless, the identical expertise SophosAI examined as a possible supply of malicious e mail and internet content material may also be used to detect that sort of content material, in addition to to categorise new web pages that don’t have any present fame knowledge. Through the use of multimodal AI that appears in any respect features of content material utilizing LLMs–fashions educated on large datasets of Web content material—to research e mail headers, content material, logos and pictures, it’s attainable to acknowledge new and beforehand unseen phishing emails and different malicious or undesirable content material even with out particular coaching knowledge.
Lee’s presentation is at 11:30 AM native time on Friday, October 4. A full article on the analysis might be revealed following the presentation.