Emad Mostak, left, of Stability.ai, joined Lightning.ai’s William Falcon on stage to highlight the essential nature of open source AI software. Tiernan Ray/ZDNET
At a celebratory meeting last Friday night on Manhattan’s West Side, two of the leading young AI software companies, Stability.ai and Lightning.ai, brought together software developers and startup owners to keep AI open source.
The hashtag for the evening is #keepAIopensource.
also: Best AI Chatbots: ChatGPT and Other Notable Alternatives
About 700 guests, including developers, representatives of major tech companies and startup owners, gathered for drinks and refreshments as the sunset streamed through the windows of the Glass House’s sixth-floor event space near the Hudson River. A lively crowd gathered around laptops on high tables to discuss demonstrations of software that uses AI algorithms in one way or another.
Banners on the display stage provide a colorful illustration of what appear to be coders happily working together under the slogan “Keep AI open source.”
Shortly before 8 p.m., Stability.ai founder and CEO Emad Mostak took the stage with Lightning.ai CEO William Falcone. The two took turns presenting their views on why AI should remain an open source affair.
also: Open source essential to alleviating AI fears, says Stability.ai Why founder
Stability.ai is best known for its Stable Diffusion service, which allows you to input an expression and turn it into any style of image.
Lightning.ai is best known for a library of functions that plug into PyTorch to solve the problem of training programs, including the large language models behind OpenAI’s ChatGPT.
The evening’s event was clearly a reference to the unexpected tilt by prominent companies to keep the technical details and source code of AI programs secret. In March, OpenAI broke industry precedent by refusing to reveal the technical details of its latest major language model, GPT-4.
About 700 guests, including developers, representatives of major tech companies and startup owners, gathered for drinks and refreshments as the sunset streamed through the windows of the sixth floor event space overlooking the Hudson River. Tiernan Ray/ZDNET
Google mimicked OpenAI this month when it refused to reveal the technical details of its new PaLM 2 language model.
Both gestures run counter to decades of revelations by AI scientists, and industry luminaries have warned that such secrecy could have a chilling effect on AI research. Mostaque warned that closed-source AI programs are incompatible with the future role of AI, and he promised to be “the leader of the gap as everyone closes.”
also: Google is following OpenAI by saying almost nothing about its new PaLM 2 AI program
Falcon introduced media messaging and lessons from the history of open source software. He noted that the company’s software builds on the success of Torch developed 20 years ago, which was later integrated with Meta’s PyTorch library. “The things that are possible today would not be possible without torch people,” Falcone said.
Falcone played the original 1984 Macintosh computer ad reminiscent of Big Brother on the stage’s projection screen, and also assembled images from Apple’s iconic “Think Different” ad campaign when the late Steve Jobs returned to the company.
also: ChatGPT’s success could lead to a damaging AI privacy swing, says AI pioneer Bengio.
“At the end of the day, there’s a lot going on,” Falcone said. “It [the AI movement] can be fully closed or fully opened.”
Commenting on OpenAI CEO Sam Altman’s comments that the code should remain secret to protect the world from bad actors, Falcone said:
Falcon ended with a photo of Jobs making a dismissive gesture at IBM, drawing applause and laughter from the audience.
Mostak, speaking at the microphone, told the audience that the fight for open source software is about a tool, AI, that empowers people to tell stories. “You have to be able to tell your stories,” he told the audience.
also: With GPT-4, OpenAI chooses privacy over disclosure
“We want to bring stability to every country so that the people in that country can control their own destiny,” Mostak said. “Because it’s the story of your life, not a panorama controlled by a few companies.”
Big language models are “so important that they will rule our lives” and therefore “shouldn’t be shut down,” Mostak said. “They should not be black boxes, because who is going to make the decision?”
Mostak predicted that the use of such programs will spread around the world, and emphasized that the population of each country should use them to convey their own history.
“Why should an entire nation not be able to create? said Mostak.
also: 3 Ways OpenAI Says We Need to Start Addressing AI Regulation
“What happens when we give one of these models to every child on earth?” He urged. “Think about it. does anyone think that every child on earth won’t have their own artificial intelligence.”
“It’s a mission to activate the potential of humanity,” Mostak said.