ai startups and funding · ai startups and funding
TypeSafe AI Emerges From Stealth With $40M Funding and Jev Model for Programmatic Logic
Published · Updated
Cited: Business Wire, Forbes, AI News
TL;DR
TypeSafe AI, a startup founded by ChatGPT co-inventor Diogo Almeida, has emerged from two years of stealth with $40 million in funding. The company introduces Jev, a System One Model designed to replace conversational AI with type-safe, parallel-sampled programmatic decisions. Jev processes unstructured inputs and outputs structured values in a single query, eliminating hallucinations and parsing overhead.
Why now
TypeSafe AI, a startup founded by ChatGPT co-inventor Diogo Almeida, has emerged from two years of stealth with $40 million in funding. The company introduces Jev, a System One Model designed to replace conversational AI with type-safe, parallel-sampled programmatic decisions. Jev processes unstructured inputs and outputs structured values in a single query, eliminating hallucinations and parsing overhead.
Agree / conflict
6x faster than conversational models. Funding: $40M raised to support development and deployment, with early developer access now open. 042 per million tokens, with no output token charges.
Takeaway
For Developers: Jev offers a faster, cheaper, and more reliable alternative to conversational AI for production code. Early access is available now. For Businesses: Automate branching logic, data extraction, and verification layers without the risk of hallucinated outputs.
TypeSafe AI’s Jev model represents a fundamental shift in how AI integrates with software systems. Unlike autoregressive models that generate text token by token, Jev uses a hardware-aware parallel sampler to evaluate all possible outputs simultaneously. This design eliminates syntactic type failures and output hallucinations by restricting outputs to predefined schemas.
The training methodology, Reinforcement Learning for Calibrated Decisions (RLCD), ensures that confidence scores correlate directly with output accuracy—a critical improvement over RLHF-based models that prioritize conversational plausibility. 1.
Real-world deployments confirm Jev’s viability in high-speed environments. A Doom bot running at 10 queries per second costs only $7 per hour, while Wikiracing tests show Jev completing traversals in fewer steps by avoiding hallucinated dead ends. Primary applications include real-time feature extraction, petabyte-scale data workflows, output verification layers, and automated branching logic.
FAQ
What makes Jev different from ChatGPT or other LLMs?
Jev does not generate text. It takes an unstructured state as input and outputs type-safe structured values in a single parallel query. This eliminates hallucinations and parsing overhead, making it suitable for deterministic code integration.
How fast is Jev compared to traditional models?
Internal evaluations show end-to-end latencies of 70–500 milliseconds, compared to 3–329 seconds for conversational frontier models. In workflow tests, Jev executed up to 193.6 times faster.
What is the cost of using Jev?
Input processing costs $0.042 per million tokens—significantly lower than standard rates of $0.20 to $10 per million tokens. Output tokens are free because Jev generates structured values without autoregressive passes.
Who founded TypeSafe AI?
TypeSafe AI was founded by Diogo Almeida, a ChatGPT co-inventor and OpenAI veteran, who developed Jev over two years in stealth.
How can developers access Jev?
TypeSafe opened early developer access on September 16, 2026, and is onboarding engineering teams from its deployment waitlist.
Sources
Canonical URL: /trend/2026/typesafe-ai-emerges-from-stealth-with-40m-funding-and-jev-model-for-programmatic
Disclaimer
Digests summarize public sources. They are not advice, forecasts, or a complete record of every trend.
This digest was compiled by Yanuki using publicly available data and trending information. The content may summarize or reference third-party sources that have not been independently verified. While we aim to provide timely and accurate insights, the information presented may be incomplete or outdated.
All content is provided for general informational purposes only and does not constitute financial, legal, or professional advice. Yanuki makes no representations or warranties regarding the reliability or completeness of the information.
This digest may include links to external sources for further context. These links are provided for convenience only and do not imply endorsement.
Always do your own research (DYOR) before making any decisions based on the information presented.