With many companies implementing GenAI and AI into their tools, software vendor Atera has been ahead of the curve and has already found incremental ways to help MSPs with their business operations.
With what the company calls AIT (artificial IT), the Tel Aviv, Israel-based vendor has seen their MSP partners becomes 10 times more efficient.
“When we start measuring the impact, we’re seeing things like 10x improvement in IT efficiency,” Atera CEO Gil Pekelman told CRN. “If you look at the numbers, on average an IT person is able to solve seven tickets per day. [With co-pilot] it goes up to 70.”
Atera is also working alongside Microsoft to bring enhanced tools to MSPs.
“What happened was at the end of 2022 Microsoft — we’re a big Microsoft shop — came to us and said, ‘Look, we have this thing, OpenAI. Why don’t you try it and see what you can do,’” he said. “We had everything built, but once we put OpenAI in there suddenly everything came together and we immediately understood that this is a game changer.ADVERTISEMENT
“It’s very, very specific,” he added. “It learns from your network, it learns from your knowledge base. For an MSP it’s a huge leap in quality of service, it’s a huge leap in efficiency and a way to increase profits and growth. Suddenly they become a super MSP.”
It also helps decrease the talent gap MSPs continuously face, he said.
“We’re pushing this because we can see that any one of our MSP partners that use AI, it just elevates them,” he said. “It’s not an incremental improvement. There is a learning curve here. You have to learn how to use it and people are skeptical.”
Pekelman spoke to CRN about what the company is doing with GenAI and what MSPs should look out for from the vendor in 2024.
Why did Atera go all in on GenAI so early?
We started working on AI, not specifically GenAI, around 2016. What we do specifically is amazing for AI because we collect data and analyze the data to predict a problem, and then we should be able to automatically fix it. It made a lot of sense to us to create this engine in the middle between the data and the solutions and automate it all in some way. But we actually never succeeded in really making that whole thing work.
What happened was at the end of 2022 Microsoft — we’re a big Microsoft shop — came to us and said, ‘Look, we have this thing, OpenAI. Why don’t you try it and see what you can do.’ We had everything built, but once we put OpenAI in there suddenly everything came together and we immediately understood that this is a game changer.
How has using GenAI in your company grown over the last 12 months?
We looked at the day-to-day jobs of our MSPs and created a co-pilot with lot of tools and capabilities that make you more efficient. That was the first phase, and once you start using it it immediately boosts your efficiency. But the more interesting phase came after that, which we’ve been rolling out slowly, is the ability to solve problems autonomously without a human being.
If the AI is not able to solve it, [it goes to] a tier one engineer. If this tier one engineer is not able to solve the problem, they then escalate it to a tier two engineer. When it’s escalated and the co-pilot meets the tier two technician, co-pilot has already done all the research for them. So it reduces the time and increases efficiency dramatically. It’s also able to run tests on your network, on your computer or on your firewall with various cloud providers and fix the problem without any need of a human technician there to help. When we start measuring the impact, we’re seeing things like 10x improvement in IT efficiency. If you look at the numbers, on average an IT person is able to solve seven tickets per day. [With co-pilot] it goes up to 70.
How do you think this leads to decreasing the skills gap?
It definitely helps the knowledge gap. One of the things that the autopilot does, besides summarizing the problem and checking it, it can offer multiple solutions. So suddenly the junior IT person has these different solutions and may say, ‘That’s one that I didn’t think of.’ It accelerates younger people earlier in their career to handle and tackle much more complex problems.
What are the biggest threats you think AI poses to IT environments?
The biggest risk is in security because you want to know that whatever you give to the AI stays contained, doesn’t spread out and that it isn’t used to learn and help other people. You will want to know that it can’t be hacked and do all kinds of things to your data. Before bringing AI into the company on any type of solution, one has to verify what kind of security is around it. You need to make sure that you’re choosing a vendor that’s secure.
How are you securing the data used to train AI models?
When we use [Microsoft] Azure OpenAI, there’s the general data that it has which is trillions and trillions of data points. Then there’s the unique data of the company itself that is using it and that is contained in a closed database. It’s private. It is actually not used to train the model but is used to prompt the model.
What is Atera’s 2024 roadmap when it comes to using AI?
The major push for the year is to increase the capabilities of autopilot. It’s autonomous but it’s limited in what it’s allowed to do. We want to increase that. The more coverage we do for the IT person, the fewer things that they have to deal with.