Vertical AI and Purpose-Built Hardware Drive Next Wave of Technology Growth

A pattern is crystallizing across the technology sector: the companies poised for substantial growth are those building AI systems designed for particular verticals, leveraging proprietary datasets and increasingly incorporating custom hardware. Participants at the Bank of America Private Tech Trailblazers Conference 2026 in Palo Alto, California demonstrated this shift across sectors ranging from food service and construction to healthcare, international payments and military applications. These organizations are concentrating on well-defined challenges where artificial intelligence can deliver quantifiable business outcomes.

As foundation models grow more commoditized, the sources of competitive differentiation are shifting. Domain-specific information, vertical integration, reputation and the financial resources needed to scale operations are becoming the decisive factors. Organizations that control larger portions of their technology stack—spanning GPU resources, custom models, robotic systems and production capabilities—are betting that long-term viability will outweigh short-term innovation.

The thing about AI is that specialized intelligence is a big story now. It was a sleeper story about a year ago, but domain-specific data, these verticals have the necessary requirements because the GPUs want data, the systems want data, the intelligence comes out of the data, and that is the key.

John Furrier, executive analyst at theCUBE Research

Nine Key Developments from the Conference

1. Bear Robotics Scales Humanoid Deployment Alongside Mobile Robot Fleet

Bear Robotics Inc. currently operates approximately 16,000 autonomous mobile robots in active service, maintains a backlog of roughly 4,000 units and achieves annual revenue doubling, according to Bren Pierce, co-founder of Bear Robotics. The company's primary focus remains restaurants across Japan and Korea, though it also serves care facilities and gaming establishments. An arrangement with LG is expanding operations into warehouse and manufacturing environments. Pierce previously founded Kinisi, a humanoid robotics startup, which he reacquired and integrated into Bear. The humanoid systems utilize the same software platform and cloud infrastructure as the mobile robot line. Three technological advances have compressed timelines from six months to just days: foundation models requiring only hundreds of examples for training, onboard processing power such as Nvidia Corp.'s Jetson Thor, and large language model-assisted development. The primary remaining obstacle involves tactile manipulation, with advanced robotic hands currently costing approximately $30,000 each.

2. Construction Automation Advances When Job Sites Become Structured

Labor expenses represent 55% to 60% of total construction expenditures. Address Robotics Ltd., operating under the name All3, is constructing an end-to-end system designed to reduce this proportion, according to Rodion Shishkov, co-founder and chief executive officer of All3. The company's software generates building designs from initial site assessment through permit-ready documentation. A manufacturing component then directs industrial robots to fabricate custom elements at mass-production economics. The Mantis mobile robot, developed by the company, performs on-site assembly and finishing work. By designing both the workflow and the robotic systems, All3 can restructure processes to eliminate most edge cases rather than programming robots to manage every scenario. Following a seed investment between $25 million and $30 million, the company is preparing its inaugural project: a six-story co-living structure on an unusually shaped 11-sided parcel.

3. Bloomreach Demonstrates Proprietary Models Outperforming Commercial LLMs

The competitive landscape for AI is shifting as model technology becomes increasingly standardized, with value concentration moving toward specialized applications in sectors like e-commerce and marketing, explained Raj De Datta, co-founder and CEO of Bloomreach Inc. The company deploys roughly 100 models. Its Loomi AI engine, built on training data from 7 billion consumer profiles, generates models delivering five to 10 times superior performance compared to standard commercial language models. Approximately half of the company's customer base has already adopted an AI agent, and the number of customers using four agents has expanded 23-fold over the past year. Revenue acceleration is occurring without corresponding increases in employee count. Through the Loomi Connect platform, external agents can directly access the system, and these interactions are expanding at an 83% month-over-month rate.

4. Electric Medium-Duty Vehicles Match Diesel Pricing While Cutting Fuel Costs

Harbinger Motors Inc. manufactures electric and hybrid medium-duty vehicle platforms priced competitively with conventional diesel options. A standard delivery truck operating in California can reduce annual fuel expenses by approximately $30,000 after accounting for charging costs, according to John Harris, co-founder and CEO of Harbinger. The company's client roster includes FedEx Corp. and Thor Industries Inc. Battery technology originally developed for delivery vehicles now supplies power systems for Airstream recreational trailers. Harbinger manufactures delivery trucks, recreational vehicle chassis, energy storage products and Army autonomous ground vehicles using shared production infrastructure, and only pursues business opportunities requiring no additional capital investment. A collaboration with American Rheinmetall targets autonomous ground systems for military applications. The company roughly doubles its production capacity annually.

5. Unconventional AI Achieves Massive Power Reduction Through Co-Designed Hardware and Software

Unconventional Inc., operating as Unconventional AI, is developing integrated hardware and software solutions to decrease AI power consumption by approximately 1,000 times, according to Ali Esfahani, chief financial officer of Unconventional AI. Naveen Rao established the company one year ago. The organization has secured approximately $540 million in funding, expanded from 14 team members in January to roughly 60, and completed a chip tape-out at Taiwan Semiconductor Manufacturing Co. on June 1. The resulting chip demonstrated power characteristics substantially superior to conventional systems. The methodology employs physics-based computational dynamics on conventional semiconductor manufacturing processes, with the system's changing state serving as memory storage, analogous to neural activity in biological brains. Esfahani contended that because electrical grid expansion follows linear growth patterns, current architectural approaches cannot sustain future demands.

6. Airwallex Achieves Rapid Growth by Simplifying Cross-Border Financial Operations

Airwallex Pty Ltd. operates a worldwide financial infrastructure enabling organizations to establish accounts, process payments and distribute cards across the world's 80 to 100 largest economies with local-market efficiency, according to Irvin Sha, head of corporate development, capital markets and investor relations at Airwallex. The company was established in Melbourne in 2015 and invested years accumulating more than 90 regulatory licenses, banking relationships and payment network agreements. The platform now incorporates AI capabilities allowing customer agents to operate autonomously. In the emerging domain of agentic commerce, which Sha characterized as a competitive race, interoperability and transaction accuracy protecting all participants will prove critical. Airwallex completed funding rounds designated Series F, G and H, raising approximately $960 million at valuations reaching $11 billion, and currently generates approximately $1.4 billion in annualized revenue.

7. Voice Agents Extend Healthcare Capacity Without Diagnostic Functions

Hippocratic AI Inc. develops voice-based agents managing appointment scheduling, pre-operative preparation, post-discharge follow-up and ongoing disease management for healthcare systems, insurance providers and pharmaceutical organizations, according to Shubhra Jain, chief business officer of Hippocratic AI. The system maintains strict boundaries against diagnosis or prescription. The safety architecture incorporates 31 models: one orchestrates conversation while 30 provide oversight. Six participating health systems that invested in the company contributed 6 million authentic patient interactions for model refinement. Because the agents operate without pressure to minimize call duration, they can cultivate relationships through conversation history maintained across multiple interactions. Over three years, the company has secured contracts with more than 60 enterprise organizations, including five of the nation's largest insurance companies.

8. Capital Requirements and Scale Reshape AI Company Trajectories to Public Markets

The most significant transformation in the sector concerns the capital intensity of AI and robotics ventures, resulting in companies reaching public market status at substantially larger scales, according to JD Moriarty, vice chairman and managing director and global head of TMT equity capital markets at Bank of America Corp. The institution monitors organizations it categorizes as AI tailwind companies. Several that faced difficulties two years prior are now considering public offerings. Public market investors prioritize sustainable, exceptional growth above other factors, and currently scale is outweighing differentiation, as CoreWeave Inc. demonstrated. During 2026, capital markets engagement concentrates on hardware and semiconductor sectors rather than software, with the bank developing comprehensive capital solutions extending beyond initial public offerings.

9. CloudWalk Achieves Near-Complete Automation of Customer Support Through Proprietary GPU Infrastructure

CloudWalk Inc. serves more than 10 million active users through InfinitePay in Brazil, its newly launched Pierre agent and JIM.com in the United States, and has exceeded $2 billion in revenue with hundreds of millions in profit, noted Luis Silva, CEO and founder of CloudWalk. Half of these users interact with the company's agents daily. The agents operate on CloudWalk's proprietary cluster containing hundreds of Nvidia Blackwell GPUs. As a regulated financial institution, CloudWalk embedded safeguards into agent architecture ensuring compliance with customer instructions and minimizing hallucinations during financial transactions. Agents currently manage 99% of customer support interactions, up from 65% eighteen months earlier. The company generates $2.7 million in revenue per employee.

https://www.youtube.com/embed/videoseries?list=PLZvOz3_Xy7sM