Cognichip raises $60M to use AI for designing the chips that power AI itself

Cognichip Secures $60 Million to Transform AI-Driven Chip Design

Cognichip emerged from stealth with a bold proposition: use AI to design the very chips that power artificial intelligence. The startup announced Wednesday it raised $60 million in new funding led by Seligman Ventures, with participation from Mayfield, Lux Capital, FPV and Candou Ventures2

2

. Intel CEO Lip-Bu Tan invested through his venture firm Walden Catalyst Ventures and will join Cognichip's board1

1

. Umesh Padval, managing partner at Seligman Ventures, will also join the board. The company has now raised $93 million altogether since its founding in 20241

1

.

Why the Semiconductor Industry Needs This Now

The problem Cognichip tackles has plagued the semiconductor industry for decades: chip design is enormously complex, expensive, and slow. Advanced computer chips take three to five years to go from conception to mass production, with the design phase alone consuming up to two years before physical layout begins1

1

. Consider that Nvidia's latest Blackwell GPUs contain 104 billion transistors—each requiring precise placement1

1

. CEO and founder Faraj Aalaei says markets can shift dramatically during these extended timelines, potentially rendering massive investments obsolete. The startup claims its technology can reduce development costs by more than 75% and cut timelines by more than half1

1

.

Physics-Informed Foundational Model Sets Cognichip Apart

Cognichip hasn't built another electronic design automation tool. Instead, it developed the Artificial Chip Intelligence platform, a physics-informed foundational model built specifically for semiconductor design2

2

. Unlike general-purpose AI models, it integrates physical constraints, circuit behavior, and manufacturing difficulties directly into the design process, enabling it to reason across every step from architecture to verification and production2

2

. Traditional chip design processes are sequential, moving step-by-step through workflows. Cognichip's approach embraces parallelism, allowing multiple design decisions to be explored simultaneously—critical when advanced chips span digital, analog and mixed-signal domains with complex interdependencies2

2

.

Overcoming Data Challenges in a Secretive Industry

The company's advantage lies in using its own model trained on chip design data rather than starting with a general-purpose LLM. But accessing domain-specific training data proved challenging. Unlike software developers who share code openly, chip designers guard their intellectual property closely, making the open-source repositories that typically train AI coding assistants largely unavailable1

1

. Cognichip developed its own datasets, including synthetic data, and licensed data from partners. The startup also created procedures allowing chipmakers to securely train Cognichip's models on proprietary data without exposing it1

1

. Where proprietary data isn't available, Cognichip uses open-source alternatives. In a demonstration last year, electrical engineering students at San Jose State University used the model to design CPUs based on the RISC-V open-source chip architecture during a hackathon1

1

.

Market Momentum and What Lies Ahead

Cognichip says it's working with more than 30 semiconductor design companies, including some of the industry's biggest players, with its platform now being tested in real-world production workflows2

2

. Early adopters report reductions in chip design cycles and costs along with enhanced performance. However, the company hasn't disclosed specific customers and can't yet point to a new chip designed with its system1

1

. The startup faces competition from incumbent players like Synopsys and Cadence Design Systems, plus well-funded startups including Alpha Design AI, which raised $21 million Series A in October 2025, and ChipAgentsAI, which closed a $74 million extended Series A in February1

1

. Padval noted that the current flood of capital into AI infrastructure is the largest he's seen in 40 years of investing, calling it a super cycle for semiconductors and companies that accelerate innovation in the space1

1

. Lip-Bu Tan stated, "The semiconductor industry is at a critical juncture; an AI framework for innovation and efficiency will unlock massive global opportunity"2

2

. If Cognichip succeeds in compressing design timelines from years into months, it could accelerate chip innovation and boost momentum across the entire AI ecosystem.