Observed Signal · Jun 25, 2026 · Technical Release · Source: techcrunch · Impact: 3/5 · Sentiment: Positive
Ex‑Databricks AI Chief Claims 1,000× Energy Cut with Oscillator Chips
Naveen Rao, formerly head of AI at Databricks, is leading Unconventional AI, a startup rebuilding compute architecture around an oscillator‑based design to drastically reduce inference energy. The company published a paper and released its first model, Un-0, an image‑generation system running on a software simulation of its oscillator chips that the team says matches state‑of‑the‑art diffusion models. Unconventional plans to publish schematics for a physical chip and to build a full inference stack, with the long‑term goal of providing inference compute at roughly 1/1000th the power of conventional systems. The company remains small (under 50 employees) and much of the hardware and infrastructure is still under development.
The announcement introduces a novel oscillator‑based compute architecture and a working simulated model (Un-0) that claims parity with diffusion image models while promising up to 1,000× lower inference energy—an innovation with potential to materially affect AI infrastructure and inference cost if validated and scaled, but it is early stage and unproven in hardware.
Track Databricks Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- Naveen Rao, formerly head of AI at Databricks, leads Unconventional AI.
- Unconventional AI released its first model, Un-0, an image‑generation system built on a software simulation of oscillator‑based chips.
- The company’s paper claims Un-0 performs comparably to state‑of‑the‑art diffusion image models.
- Unconventional AI plans to release schematics for a physical oscillator chip and build an end‑to‑end inference stack.
- The company has fewer than 50 employees and claims oscillator‑based computing could reduce inference power use by up to 1,000×.
Connected Companies & Entities
3 Entities mapped“Led by Naveen Rao, formerly the head of AI at Databricks, Unconventional AI promises to make inference processing vastly more power efficien...”
“The output from the new Un-0 model is similar to that of image-generation models like Stable Diffusion or OpenAI’s GPT Image 1....”
““This is the ‘hello world’ of a new kind of computer,” Rao told TechCrunch....”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
OpenAI’s Jalapeño Chip Shows Leading Inference Efficiency
OpenAI revealed Jalapeño — its first custom inference ASIC and rack-scale platform co-developed with Broadcom and Celestica and shown at Hot Chips — with A0 engineering samples taped out Nov 2025. The compute die uses TSMC N3P, MXFP numeric formats and HBM4 (15.4 TB/s per package); package TDP is ~700 W with sustained test power ≲ ~550 W. The rack design keeps model state (KV cache) local, simplifies on-node fabric and can scale to 2,048 XPUs. OpenAI and SemiAnalysis/InferenceX benchmarks report substantial performance-per-watt and latency gains (and partial advantages vs Nvidia GB300), but results are not independently verified and did not include Nvidia Vera Rubin. OpenAI targets small-volume deployment end‑2026 and broader ramp in 2027, pursuing a multi‑vendor production strategy while production economics, yield and fleet reliability remain unproven.
OpenAI unveils Jalapeño inference chip with Broadcom
OpenAI unveiled its first custom-built inference processor, called Jalapeño, developed in collaboration with Broadcom. The chip is designed specifically for inference workloads and, according to OpenAI, early tests show substantially better performance-per-watt than current alternatives. OpenAI said its own AI models assisted chip development. The partnership with Broadcom was announced previously in October, and the move is widely seen as a way for OpenAI to reduce reliance on Nvidia GPUs for inference, while heavier tasks like pre‑training will likely continue to use existing GPU hardware. OpenAI framed the chip as part of a broader strategy to optimize across the stack — from chip architecture to deployment systems — to make models faster, more reliable, and cheaper to run.
Etched Builds Specialized Hardware for AI Inference
This a16z opinion piece argues that AI inference is becoming the largest, most stable computing workload and therefore will favor specialized hardware over general-purpose GPUs. It highlights Etched, a startup founded by Gavin Uberti, Robert Wachen, and Chris Zhu, which is building full-stack inference systems (chips, boards, interconnects, racks) optimized for tokens-per-watt via "low-voltage inference" and "cluster-scale memory." Etched reportedly taped out first-pass silicon on TSMC's N4P process, has hired 400+ engineers from major hardware companies, and plans to ship its first racks to customers this summer. a16z says it is partnering with Etched.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.