In the years since, a wave of investment by companies large and small has spread facial recognition around the world, installed always-listening virtual assistants in homes, and led to AI technology becoming an integral part of almost every gadget, app, and service .
The race is now on to find the applications of Generative AI that will shape the world. One of the early successes is Microsoft’s Copilot, which can write code for a specific task and costs $10 a month. Another is Jasper, which offers a service that automatically generates text for businesses to use in blog posts, marketing copy, and emails. Last week, the company announced it had raised $125 million in funding from investors who valued the company at $1.5 billion and claimed to have $75 million in revenue this year.
Both Microsoft and Jasper were built on the services of OpenAI, an AI company that started as a nonprofit with funding from Elon Musk and other tech luminaries. It pioneered text generation, starting in 2019 with an algorithm called GPT-2. In late 2021, a more powerful commercial successor known as the GPT-3 was made available for everyone to use.
OpenAI has also fueled the recent surge in interest in AI imaging by announcing in January 2021 a tool called DALL-E that could generate rough images for a text prompt. A second version, DALL-E 2, released in April 2022, is capable of rendering more sophisticated and complex images, showing how quickly the technology has evolved. A number of companies, including Stability AI, now offer similar tools for creating images.
Of course, the hype surrounding Silicon Valley can be ahead of reality. “There’s a lot of FOMO,” says Nathan Benaich, an investor at Air Street Capital and author of The State of AI, an annual report that tracks technology and business trends. He says Adobe’s $20 billion acquisition of Figma, a collaborative design tool, created a sense of rich opportunity in reinventing creative tools. Benaich looks at several companies exploring the use of Generative AI for protein synthesis or chemistry. “It’s pretty crazy right now – everyone’s talking about it,” he says.
Joanne Chen, a partner at Foundation Capital and an early investor in Jasper, says turning a generative AI tool into a valuable business is still difficult. Jasper’s founders put most of their effort into fine-tuning the product to meet customers’ needs and tastes, she says, but she believes the technology could have many uses.
Chen also says the generative AI onslaught means regulation has yet to catch up with some of the unsavory or dangerous uses it might find. She worries about how AI tools could be misused, for example to create videos that spread misinformation. “What worries me the most is how we feel about safety and false and fake content,” she says.
Other uncertainties surrounding generative AI raise legal questions. Amir Ghavi, a corporate partner at law firm Fried Frank, said he recently answered a number of questions from companies looking to use the technology. They face issues such as the legal implications of using models that can be trained on copyrighted material, such as B. images scraped from the Internet.
Some artists have complained that image generators threaten to undermine human creativity. Shutterstock, a stock image provider, announced this week that it would offer an OpenAI-powered image generation service, but would also launch a fund that pays people who create images that the company licenses as training material for AI models. Ghavi says using copyrighted material to train AI models is most likely covered by fair use, which exempts it from copyright, but adds he expects this to be tested in court.
The unresolved legal issues and the potential for malicious use of generative AI do little to dampen investor interest. Their enthusiasm is reminiscent of previous Silicon Valley frenzy over social apps and cryptocurrency. And the technology at the heart of this hype cycle can help keep the speculative flywheel spinning.
Venture capital firm Sequoia Capital detailed the potential of generative AI in areas including speech synthesis, video editing, and biology and chemistry in a blog post last month. A postscript at the end noted that all images and some of the text, including future use cases for generative algorithms, were generated using AI.
