Adobe, IBM, Nvidia and five other companies have joined the initiative, unveiled in July, aimed at preventing the misuse of AI’s capabilities for harmful purposes.