Akumen AI has developed a wearable device called INSIGHT that uses cameras, sensors, and artificial intelligence to provide real-time audio and haptic assistance for blind and low-vision users, processing visual information directly into spoken guidance without requiring constant internet connection. The technology represents a shift in how assistive devices balance processing power, privacy, and practical usability for everyday independence.

What Makes This Approach Different From Existing Assistive Technology?

While tools like Microsoft Seeing AI and Google Lookout have made strides in computer vision assistance, most existing solutions struggle with real-time scene understanding, emergency response, and heavy reliance on cloud connectivity. INSIGHT takes a different architectural approach by separating the lightweight wearable interface from the computational processing layer.

The device itself functions as a camera and sensor hub that communicates with a user's smartphone for processing, rather than cramming all computing power into the glasses. This design choice has significant practical implications. Core functions such as text reading, object detection, banknote-denomination recognition, and environmental awareness are designed to run locally on a compatible smartphone wherever possible, meaning essential features work without internet access.

"Our user research indicated that many prospective users already use compatible smartphones, creating an opportunity to use the phone's computing capabilities instead of duplicating all the processing hardware within the glasses," explained Akuvan Thiyam, CEO of Akumen AI.

Akuvan Thiyam, CEO at Akumen AI

This separation of concerns reduces the glasses' weight, heat generation, power consumption, and overall cost. It also allows the company to improve supported functions through software updates without requiring users to replace the wearable hardware regularly.

What Specific Capabilities Does INSIGHT Provide?

The device's multimodal sensor integration combines visual, audio, motion, and depth data to support a range of assistive functions. Beyond basic text reading and object detection, INSIGHT includes features designed for real-world safety and navigation.

• Text Recognition: Reads printed documents and signage aloud through natural speech synthesis.
• Object and Obstacle Detection: Identifies obstacles in the user's path and provides haptic or audio alerts to prevent collisions.
• Banknote Recognition: Identifies currency denominations, addressing a practical gap in financial independence for visually impaired users.
• Fall Detection and Emergency Alerts: Monitors motion patterns to detect falls and can trigger emergency notifications when needed.
• Navigation Support: Provides contextual audio cues to help users understand their surroundings and move confidently through spaces.

The company has prioritized Indian-language support and designed feedback mechanisms to avoid overwhelming users with unnecessary information. This human-centered design philosophy emerged from direct collaboration with blind and low-vision users throughout development.

How Does the Engineering Balance Accessibility With Technical Complexity?

Building a lightweight, comfortable wearable that integrates cameras, sensors, audio, wireless communication, and power management required multiple hardware iterations. The team had to solve challenges in printed circuit board layout, thermal management, battery placement, antenna performance, weight distribution, and mechanical enclosure design.

Firmware integration proved equally demanding. Many hardware components lacked software libraries that worked reliably within the company's architecture, so the team developed device-specific drivers and integration layers from scratch. The smartphone application serves as the computational and coordination layer, receiving information from the glasses, identifying the requested function, running task-specific AI pipelines, and returning feedback with minimal delay.

"The most important accessibility decisions were to use a companion smartphone for intensive processing, prioritize offline operation for essential functions, keep the hardware architecture modular and involve intended users throughout development," Thiyam stated.

Akuvan Thiyam, CEO at Akumen AI

The company fine-tuned and quantized AI models to meet performance targets while remaining compact and efficient enough to run locally on smartphones. This required careful trade-offs among accuracy, model size, inference speed, and power consumption. Human-computer interaction design was equally critical, determining when to use speech, short audio cues, or haptic alerts, and how much information to provide without creating cognitive overload.

What Intellectual Property Protections Has the Company Secured?

Akumen AI received U.S. Patent No. 12,693,732, titled "System and method for providing real-time insights to user," on July 28, 2026. The patent protects the specific system architecture that enables a wearable device to collect multimodal sensor information, coordinate processing across edge devices and servers, and return timely contextual feedback through audio or haptic cues.

This patent covers the particular combination of wearable sensing, power management, and communication methods that distinguish INSIGHT's approach from existing assistive technologies. The protection reflects the company's focus on balancing responsible intellectual property protection with the mission of making advanced assistive technology affordable and accessible.

Why Does This Matter for the Broader Assistive Technology Landscape?

INSIGHT's design philosophy addresses a critical gap in current assistive technology. For millions of people with visual impairments, navigating bustling streets or reading printed documents remains a quiet challenge. The device's emphasis on offline functionality, lightweight design, and user-centered development offers a model for how multimodal AI can serve accessibility needs without requiring constant connectivity or expensive specialized hardware.

The company's origin story underscores the human motivation behind the technology. CEO Akuvan Thiyam was inspired by his late uncle, Dr. Ferdinand Mugusi, who was diagnosed with retinitis pigmentosa, a progressive condition that gradually took away his vision. Despite being a highly accomplished medical professor, he had to depend on assistive solutions that were often bulky, internet-dependent, and unable to provide meaningful guidance in everyday situations. Early iterations of INSIGHT were tested with him, and his feedback helped shape both the product and the company's purpose.

As multimodal AI continues to evolve, INSIGHT demonstrates that the most impactful applications may not be the most computationally complex. Instead, they thoughtfully combine visual, audio, and tactile feedback to support independence, confidence, and dignity in everyday life for users who have historically been underserved by technology.