Verizon, Atos Partner on Computer Vision at the Edge

Verizon Business is adding the Atos computer vision to its multi-access edge computing fabric, aiming to support industrial internet of things applications that include the ability to do predictive analysis of industrial process disruptions. 

By 2025, by some estimates, computer vision will be the second-biggest use of artificial intelligence in the manufacturing industries. 

source: Fractovia 

The Atos computer vision platform will analyze 180 billion data points every hour. Using this system, the engineers and operators will be able to pinpoint exactly when and where operation downtime is predicted, up to 30 days in advance, the partners say. 

The Atos platform brings AI-powered video analytics to mission critical environments,  Verizon says. Verizon hopes the Atos BullSequana Edge servers, deployed in Verizon MEC facilities, will  strengthen its 5G edge offers and unlock new use cases. 

The solution is available to customers across a variety of industries including transport, industrials, logistics and manufacturing, Verizon says.  

The Atos Computer Vision Platform enables organizations to process and analyze massive amounts of complex video and image data in real-time to automatically monitor, manage and improve working practices, security and surveillance.

The Brain and Implications for AI

21st century artificial intelligence is dominated by deep learning. 

But one of the isues with deep learning is that it’s often completely data-driven. Prior knowledge is not incorporated. Adding prior knowlege reduces computation, but adds an element of judgment or subjectivity. 

And that will be an issue. We can go faster when we incorporate what we believe we already know. That's why we have "rules of thumb."

But humans process and perceive in ways that are not strictly based on "what exists in reality" but how we conclude things are in reality. We sometimes construct it, in other words. And humans are able to understand and sort through lots of abstractions we might just call "common sense."

Algorithms cannot do this unless they are taught. The "data" does not necessarily help. It's kind of how we had issues with machine vision. "Objects" humans easily recognized often were difficult for AI to perceive. We had to embed that knowledge in the algorithms. 

It all matters for applied AI since AI is about inferences. And inferences made by humans often involve all sorts of embedded rules that make inference generation easier and more accurate. 

AI is Becoming a Feature of Computing and Connectivity

Even when artificial intelligence is more a feature than a product, AI is intertwined with computing and connectivity these days. 

Digital Realty announced the official opening of its first data center in South Korea, said to be the first carrier-neutral facility in the country. Digital Seoul 1 (ICN10) is a multi-story facility spanning 22,000 square feet and is strategically located in the northwest region of Seoul within the Sangam Digital Media City, a newly developed urban planning zone populated with technology and media companies, serving as a hub to promote South Korea's digital economy, Digital Realty says.

With 12 megawatts of capacity, the new facility offers enterprises superior connectivity with direct access to all local exchange carriers in the Korean market as well as the global ecosystem of 4,000 participants in nearly 50 metros across 25 countries.

ICN10 is also a NVIDIA-certified colocation provider of choice in South Korea as part of the NVIDIA DGX-Ready Data Center program. The facility is designed to handle Artificial Intelligence (AI) and Machine Learning (ML) workloads and analytics from NVIDIA.

AI in Mobile Radio Access Networks Gets Results, Firm Claims

Artificial intelligence will be applied to manage mobile radio access networks, it is safe to say. 

“A leading communications service provider in China achieved an 18 percent improvement in downlink throughput ratio using Capacity Turbo in a trial in Guangzhou run across a 1755 cell, 580 base station network,” says Capacity Turbo. “Capacity Turbo also enabled the same CSP to achieve an 81 percent improvement in downlink packet loss and a 22 percent improvement in uplink packet loss across the same network.”

“In another example, the CSP used Capacity Turbo to optimize VoLTE performance and achieved an 81 percent reduction in packet loss while reducing uplink packet loss by around 22 percent,” the company says.

A major CSP in Thailand was able to improve coverage as well as increase base station throughput by between 13 percent to 15 percent, the firm claims.

Also, a CSP in Spain experienced a 15 percent improvement in average user downlink throughput across an optimized 766 cell portion of its network.

OpenAI Gets More Investment

Microsoft-backed OpenAI, the artificial intelligence lab that competes with Alphabet’s DeepMind and Meta AI, says it has raised $250 million in new investment. Microsoft invested $1 billion in 2019 into OpenAI. 

As part of that deal Microsoft become the firm’s sold cloud computing supplier. 

“After our pre-friends-and-family round in 2016, our F&F round in 2017, our angel round in 2018, our pre-seed round in 2019, our seed round in 2020, and our seed extension in 2021, we’re delighted to share we’ve raised a Series A of $250 million,” says Sam Altman, OpenAI’s co-founder and CEO.

Some might say there is an artificial intelligence investing bubble. Others will argue the potential returns are so great the “over-investment” is a rational bet. 

AI Provides Greater Cost Reduction Than Revenue Enhancement Benefits in 2020

A McKinsey survey of global enterprise executives suggests they believe artificial intelligence is boosting bottom line results by about five percent. Some 27 percent of respondents agree with that claim. 

source: McKinsey 

And while AI’s revenue benefits have held steady or even decreased since the previous survey—especially for supply-chain management, where AI was unlikely to compensate for the pandemic era’s global supply-chain challenges—the opposite is true of costs, McKinsey says. 

Respondents report significantly greater cost savings from AI than they did previously in every function, with the biggest year-over-year changes in the shares reporting cost takeout from using AI in product and service development, marketing and sales, and strategy and corporate finance.

Who Provides AIOps Now, and How Will that Change Over Time?

Analysts and researchers always must make decisions on scope when studying markets of any sort. Which organizations and firms are included within the data set, and what are the selection criteria? 

The AIOps market, generally considered to consist of information technology operations management by using machine learning or artificial intelligence, typically focuses on suppliers of such software capabilities. 

Omdia highlights ServiceNow, Digitate, BMC, IBM and Splunk, LogicMonitor, Sumo Logi, PagerDuty, StackState, VuNet Systems and Netreo as firms it can include within the AIOps market. 

source: Omdia 

Other observers might say the suppliers include :

AppDynamics

BigPanda

BMC

DataDog

DynaTrace

Moogsoft

New Relic

PagerDuty

ScienceLogic

Splunk

source: ITPro 

And yet others--using different criteria-- would add as many as 50 or so additional firms supplying AIOps capabilities defined to include Cloud Management Platforms, Privileged Access Management (PAM), IoT Platforms, Cloud Migration Tools or Hybrid Cloud Storage. 

Researchers also have to make decisions about whether AIOps is a feature of another product sold by a company, or a primary product sold by a company. Eventually, one might well argue, every enterprise software system will include the use of ML or AI to provide operating information, just as all enterprise software will use cloud delivery mechanisms. 

That virtually guarantees there will be changes in market boundaries over time. AIOps will simply become part of the capability set for all enterprise software with performance management requirements. 

So AIOps might be redefined as general purpose tools that are extrinsic to any single application or process. To the extent that ML or AI-assisted analysis and monitoring tools are supplied with any particular app, only suppliers of third party tools that monitor across domains might come to be seen as “true” suppliers of AIOps.