Visitors to events like TechEx North America will always want to see the cutting edge front and center stage, but the nuances and details brought to the show by speakers and exhibitors mean that, at least in the minds of corporate decision makers, they are sometimes small considerations that need to be weighed.
The question in fields as diverse as edge computing, IoT, data center congress, and cyber security has been: What needs to be built around AI before it is introduced into the physical, business-oriented world?
The Edge Computing track has its roots in traditional industry, examining latency, deployment discipline, and cybersecurity at the convergence of IIoT/IT. The day-one program positioned edge computing as a place where businesses can reassess the value of their data assets, consider how decisions are made by autonomous devices, and the processing speed they require.
Sessions explored scaling edge deployments (e.g., multi-site businesses), agent network operations, immutable edge infrastructure for distributed inference (on-premises, cloud, hybrid), and how lessons from zero trust cybersecurity can be applied to control systems.
Ed Doran of the Edge AI Foundation chaired the program, which started with the edge being a demanding place to operate. The track featured representatives from Akamai, Spectro Cloud, Scylos, TÜV Rheinland, OPC Foundation, and Germany’s Schneider Electric. Discussions covered manufacturing and IoT issues, and also delved into industrial automation and connected control and damping devices.
Bringing intelligence closer to the machine may change the risk profile (though in which direction it goes has been debatable), and faster local decision-making may reduce latency and dependence on central cloud services, but where are observability and control on the minds of decision makers?
The IoT Tech Expo’s opening track on Industrial IoT and Digital Twins focused on manufacturing with sessions covering smart factory trends, AI beyond Industry 4.0, asset management, a practical roadmap to escape pilot purgatory (more on that below), physical AI in everyday operations, and digital twins.
Similar to the debate around AI adoption in the knowledge sector, the gap between demonstration and adoption was the area that received the most scrutiny. Both industrial AI and back office AI may work well for presentations, but they can grind to a halt when they encounter older machines (or legacy software).
The alliterative pilot purgatory occupied considerable importance in several sessions on the various presentation stages and exhibition halls on the opening day. Rockwell Automation and Ford’s session on Physical AI and Connected Asset Intelligence was particularly focused on scaling projects that seem to work well in concept but can fail when deployed in the real world. How do you incorporate intelligence into daily operations without it becoming just another dashboard that no one owns?
Digital twins received similar recognition. A better version of the digital twin would not be a visual replica of what the devil would use, but the devil has uses. Instead, several speakers called for and presented operating models that could actually support factories, cities, and municipal facilities. In addition to pre-testing decisions and improving maintenance, what should modern digital twins be designed to accomplish?
The TechEx program connected ideas between speakers from a variety of program areas, including Siemens, South Korea’s LG CNS, and Boston Dynamics. We’ve learned throughout that smart systems need to be designed around the people and machines they’re designed to benefit from, whether they’re deeply embedded in engineering sites or back offices.
Sessions on Day 1 of the Data Center Congress track looked at the big issues facing the sector today, including construction, power, procurement, cooling, water, and the network spine needed for AI DCs. Keynote speakers and roundtable guests talked about construction disruptions and power issues, and early event attendees heard from TechEx host city Santa Clara about its own data center efforts.
DC issues remain at the center of broader debates about AI. AI as a technology relies on computing and its high density computing. This depends on power, cooling, land, and permits. A recurring theme in the infrastructure-focused consultations was how the economics of AI will impact the infrastructure stack, with the former changing rapidly and the latter taking years to mature.
TechEx events are unique in many ways because they bring together issues that affect the entire industry under one roof. A place where you can visualize the big picture. At the Data Center Conference, we learned that water and power constraints can defeat the rhetoric around AI scale. Sessions on AI and Big Data helped to allay the “runaway” mentality of AI productivity by citing unique reasons why unplanned and chaotic technology adoption is not a good fit for modern enterprises. Data centers are now one place where AI strategies become physical. The considerations for corporate boardrooms are pragmatic.
The Cyber
Sessions on shadow AI and data breaches were particularly relevant to the broader event. Staff at many companies are using AI services within business workflows, sometimes without authorization, and usually without the ability to record activity. So data governance and cyber governance are effectively the same story.
The benefits of one conference hosting complementary tracks have been demonstrated in several cases. For example, the cybersecurity track’s concerns about legacy systems were also reflected in the IoT and edge stages, raising questions about the convergence of old plant systems with modern smart intelligence. Security can be an afterthought in any situation, but in critical infrastructure like transportation and energy, cybersecurity must take center stage.
TechEx North America’s opening day track related to infrastructure gave the conference a sense of reality, at least in some ways. While AI may be discussed in terms of agent automation, deployment depends on networks, data center capacity, and cybersecurity. The Edge and IoT session showed how intelligence reaches machines and how it needs to be applied carefully and thoughtfully. A session focused on data centers showed the material limitations of physical structures, while a session on cybersecurity showed how the desire for speed can be the enemy.
The day showed thousands of attendees that bringing AI into production is more than just flipping a switch on software. We depend on everyday things like buildings, power grids, networks, and security. Companies that understand these issues are more likely to successfully implement the latest technology. The purpose of this event is to understand the big picture.
(Image source: TechForge)
Want to learn more about AI and big data from industry leaders? Check out the AI
AI News is brought to you by TechForge Media. Learn about other upcoming enterprise technology events and webinars.
