Owen Sakawa, founder of Aloe AI, wants his platform to become an “immune system for AI” and “antivirus for any AI agent.”
The idea is to add a layer to companies’ LLM that checks for bias, hallucinations, errors, compliance issues, misinformation and unsafe outputs, Sakawa said in an interview days before the TechCrunch Disrupt conference, where Elo is a top 20 finalist in the AI startup Battlefield competition.
Sakawa (pictured above) said, “AI is developing at a very fast pace, and it’s moving so fast without guard rails, without safety nets, without any mechanisms to keep it from derailing.”
Aloe AI is an API or SDK, a module that sits on top of the output layer of an AI model, “an infrastructure on top of your LLM pipeline,” as Sakawa explained. “And it basically fact-checks every single response.”
As Sakawa said, a startup’s system has its own layers or “anchors.”
The first anchor fact-checks the LLM response against verifiable sources. Then, the second anchor checks if the output violates any regulations, such as the US health privacy law HIPAA, the European far-reaching data protection and privacy law GDPR, or if it exposes some personal private information (PII). The final anchor is an audit trail that shows how all previous decisions were made and allows regulators or anyone auditing the system to “analyze the train of thought from where it made the decision to the source of that decision, the confidence score of all those decisions,” according to Sakawa.
Explicitly, Sakawa said that Aloe AI is not built on LLMs, because in his opinion, making LLMs investigate other LLMs is just putting a “Band-Aid on another wound.” However, Elo AI’s system uses AI technologies like machine learning. And there are humans in the loop, too: Allo AI’s employees, who must comply with new regulations on data protection and user protection, Sakawa said.
techcrunch event
san francisco
,
October 27-29, 2025
If you want to learn more about Elo AI – as well as check out dozens of other companies, listen to their pitches, and listen to guest speakers on four different stages – join us at Disrupt October 27 to 29 in San Francisco. Learn more here.
