In the years since, a wave of investment from companies large and small has spread face recognition around the world and has installed always-listening virtual assistants into homes.
The race is on to find applications of generative artificial intelligence that will make a difference. Microsoft's Copilot costs $10 per month and can be used to write code for a given task. Jasper is a service that auto-generates text for companies to use in marketing copy and email. The company announced last week that it had raised $125 million in funding and was on track to bring in $75 million in revenue.
Both Microsoft and Jasper built on top of services from Openai, an artificial intelligence company that began as a nonprofit with funding from Musk and others. The GPT-2 is the first text generation method. GPT-2 was thrown open for anyone to use late in 2021.
DALL-E, a tool that could produce crude images for a text prompt, was announced by Openai. The second version of DALL-E is able to render more sophisticated and complex images. A number of companies now offer the same tools.
It is possible to get ahead of reality in Silicon Valley. Nathan Benaich is an investor at Air Street Capital and the author of The State of Artificial Intelligence. The acquisition of Figm a, a collaborative design tool, by Adobe has created a sense of rich opportunities in reinventing creative tools. Several companies are looking at using generative artificial intelligence for chemistry. Everyone is talking about it at the moment.
It is not easy to turn a generative artificial intelligence tool into a valuable company according to a partner at Foundation Capital. She believes the technology could have many uses, despite the fact that most of the effort was put into fine tuning the product to meet customer needs and tastes.
The generative artificial intelligence rush means that regulation hasn't caught up with some of the potentially dangerous uses it could find. She is concerned about how the use of artificial intelligence could be used to spread misinformation. She says she is most concerned about how we think about security.
Legal questions are raised about other uncertainties. The corporate partner at Fried Frank says he has recently received a lot of questions from companies interested in using the technology. The legal implications of using models that are trained on copyrighted material have been a problem.
Artists complain that image generators undermine creativity. Shutterstock, a stock imagery provider, this week announced it would offer an image generation service powered by Openai but would also launch a fund that pays people who make images that the company licenses as training material for artificial intelligence models. The use of copyrighted material to train artificial intelligence models is most likely covered by fair use and will likely be tested in court.
The open legal questions don't seem to be slowing investors' interest They evoke previous Silicon Valley frenzies over social apps andcryptocurrencies. Technology at the center of the hype cycle can help keep the speculative flywheel spinning.
The potential of generative artificial intelligence was laid out in a recent post by the venture capital firm. All of the images and text were generated using artificial intelligence, according to a postscript.
All Rights Reserved © 2024
