I honestly believe that they closed things up not because of concerns about "safety and security" but because it was the most profitable thing to do. Other groups are publishing models that are just as good (maybe with a bit of lag compared to OpenAI), and OpenAI seems to have gutted their own safety teams.
The fact that OpenAI removed their ban on military use of the models[1] seems to be a sign that security and safety aren't the highest concern.
Any time I hear vague corporate statements like that I like to play a game of "what words did they leave unsaid"?
>>Today this includes publishing most of our AI research, but we expect that safety [of our profits] and [job] security concerns will reduce our traditional publishing in the future
