News

Malicious LLMs Are A Thing

“Mallas” may sound like innocent shorthand for Kamala Harris supporters or marshmallow lovers, but it actually describes malicious LLMs (large language models).   Academics at Indiana University Bloomington found 212 such systems on underground marketplaces, using a mix of open source and commercial  platforms.  If there’s any good news, it’s that Malla-created phishing content was easier to spot than human-created phishing content.

More News

Next Article

B2B Marketing Lacks Behavior-Based Personalization: ON24

September 11, 2024

Nearly all (99%) of B2B marketers do some type of personalization, says this ON24 report. But don’t be too impressed: most is based on static data such as account segment (52%), named accounts (46%), or persona (38%).  Barely over one-third use behaviors such as content requests (37%), contact engagement (35%), and account engagement.  The vast majority (88%) plan to use AI to help their personalization programs.

CDPI Newsletter
Previous Article

Salesforce Launches Industry-Specific AI Capabilities

September 10, 2024

Salesforce’s Dreamforce mega-conference is next week and the company’s publicity elves are already working overtime.  In the first of what are sure to be daily announcements this week, the company announced 100+ industry-specific prompts, data models, and AI capabilities across its fifteen industry clouds.  It also launched an AI Use Case Library to house its industry-specific AI offerings.

CDPI Newsletter