😺 Watch: Elorian wants to fix AI's toddler vision
The Neuron · Matthew Robinson · 2026-06-17
The Neuron profiles Elorian, an AI research lab co-founded by former Google Brain and DeepMind researcher Andrew Dai with $55M in funding, which is building models capable of genuine visual reasoning to overcome the inability of AI coding agents to accurately perceive and evaluate the visual output of their own work.
Appears in
Extraction
Topics: visual-reasoningagentic-aicomputer-visionai-startupsai-coding-agents
Claims
- Current AI coding agents cannot reliably perceive and evaluate the visual output they produce, creating a critical bottleneck in agentic software development.
- Visual reasoning—not code generation—is the key remaining barrier to AI agents fully solving programming tasks end-to-end.
- Frontier models can describe images but fail at tasks requiring tracking of multiple simultaneous visual relationships, which elementary school children solve trivially.
- Elorian is building models to reason through spatial relationships, physical constraints, and design intent rather than translating images into text and reasoning over text.
- Elorian raised $55M from Striker Ventures, Menlo Ventures, and Altimeter, with participation from prominent AI researchers including Jeff Dean.
Key quotes
When it opens the browser, looks at the interface it made, and misses the obviously broken layout you spotted in half a second? That's a critical flaw in the design of the system. Even worse, it lies to your face about how 'it looks totally fixed now!'
The road to actually 'solving' programming probably runs through something more basic: visual reasoning.
AI coding agents will keep getting better at code. The next leap is giving them enough visual understanding to inspect the thing they built, notice what is broken, and reason through the fix the way a human would.