{"id":85854,"date":"2024-10-16T16:00:00","date_gmt":"2024-10-16T16:00:00","guid":{"rendered":"https:\/\/one.sightlinemg.com\/navytimes\/uncategorized\/2024\/10\/16\/primerai-introduces-near-zero-hallucination-update-to-ai-platform\/"},"modified":"2026-08-08T01:12:55","modified_gmt":"2026-08-08T01:12:55","slug":"primerai-introduces-near-zero-hallucination-update-to-ai-platform","status":"publish","type":"post","link":"https:\/\/one.sightlinemg.com\/navytimes\/news\/your-navy\/2024\/10\/16\/primerai-introduces-near-zero-hallucination-update-to-ai-platform\/","title":{"rendered":"PrimerAI introduces \u2018near-zero hallucination\u2019 update to AI platform"},"content":{"rendered":"\n

An artificial intelligence company that contracts with the U.S. government says its AI system\u2019s analysis of large data sets can produce nearly flawless results.<\/p>\n\n\n\n

PrimerAI announced<\/a> on Oct. 14 that an update to its AI platform can achieve a near-zero hallucination rate, a result that the company believes has broader implications for the Defense Department and defense industry as a whole.<\/p>\n\n\n\n

The term \u201challucinations\u201d <\/b>in AI parlance refers to models spitting out incorrect results.<\/p>\n\n\n\n

\u201cIn high-stakes environments where precision and time lines are crucial, Primer\u2019s enhanced platform emerges as a game changing solution,\u201d the company\u2019s press release said.<\/p>\n\n\n\n

PrimerAI CEO Sean Moriarty explained what\u2019s important about the change to the system in a phone interview with Military Times.<\/p>\n\n\n\n

While many AI platforms experience a hallucination rate of 10%, Moriarty said, PrimerAI had whittled it down to .3%.<\/p>\n\n\n\n

\"\"
A screenshot of the PrimerAI interface that demonstrates the retrieval augmented generation verification process. (Cindy Ma\/PrimerAI)<\/figcaption><\/figure>\n\n\n\n

The biggest boon of the update was the ability to fact-check its own results.<\/p>\n\n\n\n

This proprietary system, which the company says captures over 99% of errors before they reach users, is called the retrieval augmented generation verification system. Large language models, or LLMs, which are AI systems that can understand and process human language, already use retrieval augmented generation when given a prompt. ChatGPT is a prime example.<\/p>\n\n\n\n

What makes PrimerAI\u2019s system novel is that once it generates a response or summary, it generates a claim for the summary and corroborates that claim with the source data, according to Cindy Ma, senior product manager at PrimerAI.<\/p>\n\n\n\n

This extra layer of revision leads to exponentially reduced mistakes, said Ma, who provided an in-person demonstration of the system at the Association of the United States Army\u2019s annual conference in Washington, D.C., <\/b>on Oct. 14.<\/p>\n\n\n\t