AI is one of the most significant trends this decade. These systems revolutionized the user interface by reversing the dynamic that required users to learn how to work with computers to one where the computers learned how to work with users. Once mature, these platforms should massively reduce the time needed to train users on new systems while increasing the effectiveness of those systems, the timeliness of the information, and the answers these systems subsequently provide.
Initial AI systems tended to be proprietary, expensive, difficult to train, and often disappointing in use. The future is cloud-based AI solutions using software like NVIDIA’s Ai Enterprise software suite running on something like VMware’s vSphere. Thas it just what NVIDIA and VMware announced this week: an One Suite AI package for the enterprise with a blend of tools and frameworks optimized for a broad range of industries including manufacturing, logistics, financial services, retail, and healthcare.
This blended solution should in turn result in individual AI applications uniquely targeting critical business needs in the target industries initially and a growing number of other industries eventually.
Let’s talk about the impact of this partnership between NVIDIA and VMware running on uniquely tuned near bare-metal VIDIA-Certified systems from companies like Dell Technologies, HPE, Lenovo, and Supermicro initially.
The Impact Of Broadly Available AIs
While we aren’t yet at a point where general-purpose AIs are practical, this effort should provide broad access to AI solutions focused on individual industries.
AI in Manufacturing
In manufacturing, you would get AIs that could monitor quality on the line, and more cost-effectively, do broad quality checks while products are being manufactured. Not only would they be able to point out defects, but also automatically begin the remediation process resulting in better line uptime and fewer products released into the defective market.
Often quality control systems only check on the process that occurred before the test, which can be problematic if that process breaks a previously approved component. When I worked at IBM, we once had a production line with 100% quality assurance based on these point tests and a 100% failure rate at the end of the manufacturing process because of this QC approach. An AI can do a far more complete job of testing the assembled system at each checkpoint, effectively preventing the problem of a late process breaking an earlier installed component.
With logistics, AIs can help certify suppliers, provide early warnings (if the suppliers are instrumented) to product shortages, and provide direction on how to offset a failed supplier or critical lost shipment. With adequate capacity and training, an AI can not only identify problems but recommend and even implement solutions early that could prevent missing delivery dates. Also, they can flag suppliers who are providing substandard parts or flag any part of the supply chain for performance or integrity issues as long as the related process is well-instrumented and tied into the AI.
Financial Services and AI
Financial services could use AIs for customer interaction while fooling the customer into thinking they are talking to a natural person. Natural Language interfaces have advanced dramatically over the last decade, and advanced AI systems like Watson have appeared so natural to customers that customers have tried to ask them out on a date. This personal touch has a real person’s performance, often even better because AIs don’t lose their temper, have a bad day, or make inappropriate remarks. I anticipate a future where significant traders get their own custom AI supplied by their broker, allowing them to anticipate trends better and increase income and avoid catastrophic outcomes.
Also read: How AI Might Change the BI Experience
Retail and AI
Retail AIs will help aggregate products so that a person shopping for one is more likely to buy adjacent products that meet their buying needs. Coupled with mixed reality, they could also put the user through virtual fashion shows and virtually show what the buyer would look like in a new outfit.
Beyond that, buyers who work with the retailer for a long time will gain the benefit of an AI that knows their taste and recommend products that, had the buyer known about them, they would have already purchased. This solution would result in better revenue for the retailer and better satisfaction and loyalty in customers.
Healthcare and AI
Finally, with healthcare, everyone is different, yet today we are often treated like we are the same because medical professionals can’t become experts on every patent they have. But one advantage to AIs is they can scale to everyone, and AIs tied to future patents should be better able to anticipate future medical needs and make sure the tests the patent has to take are uniquely targeted at those needs. Hence, you were less likely to be under or over-tested. This focus increases the timeliness and accuracy of the diagnosis while potentially significantly lowering the cost of treatment.
The Dawn of Ubiquitous AI
Partnerships like the one forged between NVIDIA and VMware promise more capable and more numerous focused AIs that could significantly improve customer/vendor relations and improve the quality of the products and services enhanced by these AIs. It paves the way to a future where significant industries are heavily supplemented with AIs; mistakes are less common because the related decisions are more often based on data, not gut.
We are entering a future where we no longer have to learn how to use technology, and instead, technology has to learn how to work with us; this announcement made that outcome more certain and more near term. That is the promise of ubiquitous AI and NVIDIA, and VMware just got us a ton closer to that ideal.
Read next: How AI Will Be Pushed to the Very Edge