Majority of the developers are familiar with their code. Fewer understand what their code relies on – and such gap is the place where modern precarities abide. One out of date library three levels deep in a dependency tree will…
The vast majority of engineering teams use GitHub more like a file cabinet, with its contents organized well up, instantiating code, setting up a pipeline, and leaving the settings page mostly untouched. Such a solution proved rational when GitHub was…
Have you forgotten being able to simply ship code and not document all the components within it? Those days are gone. I began to take an interest in Software Bill of Materials (SBOM) after completing a project that was flagged…
In December 2020, at the time of the SolarWinds breach, I recollect that I thought it was another advanced hack. However, getting into the technical report and observing the way how offenders changed the playbook in the following five years,…
I understand, of course, that when you hear the term software supply chain security you think it is another buzzword of the salespeople. However, since SolarWinds made it apparent that a single breach of a build system can take down…
AI agents no longer simply answer questions, and instead, they are booking meetings, changing databases, and transferring money. That change that was made on the former of reaching the point of a helper to an independent worker alters all on…
Ey, independent AI agents are no longer fancy chatbots. They are making decisions, making APIs and coordinating workflows without having human oversight at all times. Bringing, but at the same time, that requires us to have stringent governance systems into…
Well, I have spent a fair amount of time as an agent system hacker, and I can tell that they are not as impregnable as the hype they develop can make them seem. Once you put two or more AI…
And today AI agents are ubiquitous. They are responding to the customer support tickets, writing code, maintaining workflows, and making a decision that would otherwise have to be done by a human being. However, the point is here no one…
By the way, I used the last few months to test various tools based on LLM, and something kept bothering me. Why then did some of the AI agents abruptly begin to act strange after some sessions? It turns out…
Feel free to believe me the first thing that crossed my mind when I was told about the concept of memory poisoning in AI systems was a matter of science fiction. After which I began researching the specifics of how…
What Makes Autonomous Agents Vulnerable Autonomous agents are not text generative chatbots. They are AI systems that are actually powerful, which is that they can send email messages and query databases and run code and browse the Internet and can…