Earlier this year, LinkedIn co-founder and venture capitalist Reid Hoffman issued a warning mixed with amazement about AI. "There is literally magic happening," said Hoffman, speaking to technology executives across sectors of the economy.

Some of that magic is becoming more apparent in creative spaces, like the visual arts, and the idea of "generative technology" has captured the attention of Silicon Valley. AI has even recently won awards at art exhibitions.

But Hoffman's message was squarely aimed at executives.

"AI will transform all industries," Hoffman told the members of the CNBC Technology Executive Council. "So everyone has to be thinking about it, not just in data science."

The rapid advances being made by Copilot AI, the automated code writing tool from the GitHub open source subsidiary of Microsoft, were an example Hoffman, who is on the Microsoft board, directly cited as a signal that all firms better be prepared for AI in their world. Even if not making big investments today in AI, business leaders must understand the pace of improvement in artificial intelligence and the applications that are coming or they will be "sacrificing the future," he said.

"100,000 developers took 35% of the coding suggestions from Copilot," Hoffman said. "That's a 35% increase in productivity, and off last year's model. ... Across everything we are doing, we will have amplifying tools, it will get there over the next three to 10 years, a baseline for everything we are doing," he added.

Copilot has already added another 5% to the 35% cited by Hoffman. GitHub CEO Thomas Dohmke recently told us that Copilot is now handling up to 40% of coding among programmers using the AI in the beta testing period over the past year. Put another way, for every 100 lines of code, 40 are being written by the AI, with total project time cut by up to 55%.

Copilot, trained on massive amounts of open source code, monitors the code being written by a developer and works as an assistant, taking the input from the developer and making suggestions about the next line of code, often multi-line coding suggestions, often "boilerplate" code that is needed but is a waste of time for a human to recreate. We all have some experience with this form of AI now, in places like our email, with both Microsoft and Google mail programs suggesting the next few words we might want to type.

AI can be logical about what may come next in a string of text. But Dohmke said, "It can't do more, it can't capture the meaning of what you want to say."

Whether a company is a supermarket working on checkout technology or a banking company working on customer experience in an app, they are all effectively becoming software companies, all building software, and once a C-suite has developers it needs to be looking at developer productivity and how to continuously improve it.

That's where the 40 lines of code come in. "After a year of Copilot, about 40% of code was written by the AI where Copilot was enabled," Dohmke said. "And if you show that number to executives, it's mind-blowing to them. ... doing the math on how much they are spending on developers."

With the projects being completed in less than half the time, a logical conclusion is that there will be less work to do for humans. But Dohmke says another way of looking at the software developer job is that they do many more high-value tasks than just rewrite code that already exists in the world. "The definition of 'higher value' work is to take away the boiler-plate menial work writing things already done over and over again," he said.

The goal of Copilot is to help developers "stay in the flow" when they are on the task of coding. That's because some of the time spent writing code is really spent looking for existing code to plug in from browsers, "snippets from someone else," Dohmke said. And that can lead coders to get distracted. "Eventually they are back in editor mode and copy and paste a solution, but have to remember what they were working on," he said. "It's like a surfer on a wave in the water and they need to find the next wave. Copilot is keeping them in the editing environment, in the creative environment and suggesting ideas," Dohmke said. "And if the idea doesn't work, you can reject it, or find the closest one and can always edit," he added.

The GitHub CEO expects more of those Copilot code suggestions to be taken in the next five years, up to 80%. Unlike a lot going on in the computer field, Dohmke said of that forecast, "It's not an exact science ... but we think it will tremendously grow."

After being in the market for a year, he said new models are getting better fast. As developers reject some code suggestions from Copilot, the AI learns. And as more developers adopt Copilot it gets smarter by interacting with developers similar to a new coworker, learning from what is accepted or rejected. New models of the AI don't come out every day, but every time a new model is available, "we might have a leap," he said.

But the AI is still far short of replacing humans. "Copilot today can't do 100% of the task," Dohmke said. "It's not sentient. It can't create itself without user input."

With Copilot still in private beta testing among individual developers 400,000 developer signed up to use the AI in the first months it was available and hundreds of thousands of more developers since GitHub has not announced any enterprise clients, but it expects to begin naming business customers before the end of the year. There is no enterprise pricing information being disclosed yet, but in the beta test Copilot pricing has been set at a flat rate per developer $10 per individual per month or $100 annually, often expensed by developers on company cards. "And you can imagine what they earn per month so it's a marginal cost," Dohmke said. "If you look at the 40% and think of the productivity improvement, and take 40% of opex spend on developers, the $10 is not a relevant cost. ... I have 1,000 developers and it's way more money than 1000 x 10," he said.

The GitHub CEO sees what is taking place now with AI as the next logical phase of the productivity advances in a coding world he has been a part of since the late 1980s. That was a time when coding was emerging out of the punch card phase, and there was no internet, and coders like Dohmke had to buy books and magazines, and join computer clubs to gain information. "I had to wait to meet someone to ask questions," he recalled.

That was the first phase of developer productivity, and then came the internet, and now open source, allowing developers to find other developers on the internet who had already "developed the wheel," he said.

Now, whether the coding task is related to payment processing or a social media login, most companies whether startups or established enterprises put in open source code. "There is a huge dependency tree of open source that already exists," Dohmke said.

It's not uncommon for up to 90% of code on mobile phone apps to be pulled from the internet and open source platforms like GitHub. In a coding era of "whatever else is already available," that's not what will differentiate a developer or app.

"AI is just the third wave of this," Dohmke said. "From punch cards to building everything ourselves to open source, to now withina lot of code, AI writing more," he said. "With 40%, soon enough if AI spreads across industries, the innovation on the phone will be created with the help of AI and the developer."

Today, and into the foreseeable future, Copilot remains a technology that is trained on code, and is making proposals based on looking things up in a library of code. It is not inventing any new algorithms, but at the current pace of progress, eventually, "it is entirely possible that with help of a developer it will create new ideas of source code,," Dohmke said.

But even that still requires a human touch. "Copilot is getting closer, but it will always need developers to create innovation," he said.

Continue reading here:

Microsoft's GitHub Copilot AI is making rapid progress. Here's how its human leader thinks about it - CNBC