Two former OpenAI safety researchers say they were fired for putting safety ahead of the company's short-term interests, and that independent safety audits are not going well, according to a report by NPR. The report, which is based on the researchers' first interview since their dismissal, is available in full at NPR's report on the fired researchers.

According to NPR, the researchers, Mikita Balesni and Tomek Korbak, spoke in their first interview since being fired from the company. NPR's summary says the pair state that they were terminated for prioritizing safety over the short-term interests of the company. The summary identifies both as OpenAI safety researchers.

NPR's headline states that the researchers' view of independent safety audits is that they "aren't going well." The title and summary do not describe how the audits were carried out, who conducted them, or what they found, so those details are not reported in this account.

NPR's summary does not say when the two researchers were fired, nor does it describe the terms of their departure. The account rests on the researchers' own statements about why they were terminated, which NPR attributes to them.

The summary also does not say whether OpenAI has responded to the researchers' statements. No other outlet's account of the matter was available for comparison, so the claims are reported as the researchers' own, as attributed by NPR. Any response from the company would need to come from separate reporting.

NPR's story was written by Kevin Roose and describes the interview as the first either researcher has given since being fired. Further details of the reporting are not contained in the material used for this account.

Sources: NPR: Fired OpenAI researchers say independent safety audits aren't going well.