Is Hyperscience profitable?

Hyperscience does not disclose profitability. It reports 37% year over year revenue growth for FY2024 (ended Feb 28, 2025) with 90%+ gross margin, per company statements from May 5, 2026, but no net income or profit figures are published since it is a private company.

Has Hyperscience had layoffs?

Yes. In March 2022 Hyperscience cut about 100 employees, roughly 25% of staff, coinciding with CEO Peter Brodsky's departure. Headcount has not recovered since, standing at about 215 employees in June 2026, down 16% year over year from a peak of 347 in mid 2021.

Could AI replace Hyperscience?

Frontier multimodal models like GPT-4o, Gemini, and Claude can already perform zero-shot document extraction, commoditizing basic IDP use cases as of February 2026. Hyperscience is repositioning around FedRAMP High government compliance and an orchestration layer (Inference Layering Optimization, April 2026) rather than raw extraction to stay differentiated.

Who owns Hyperscience?

Hyperscience is privately held and venture backed, with no parent company. Investors include Bessemer Venture Partners, Battery Ventures, FirstMark, Stripes, and Tiger Global, as of a June 17, 2025 partnership announcement. Its last disclosed funding round was a $100M Series E in December 2021 valuing it around $1.65 billion.

Who is Hyperscience's biggest competitor?

Gartner's inaugural Magic Quadrant for Intelligent Document Processing, published September 3, 2025, named Hyperscience a Leader alongside ABBYY, Rossum, Tungsten Automation, and UiPath's Document Understanding module as co-Leaders, though none has publicly taken Hyperscience's government contract share.

Has Hyperscience raised new funding recently?

No. Hyperscience's last disclosed funding round was a $100M Series E in December 2021 at approximately $1.65 billion valuation. As of July 21, 2026, no new round has been reported in over 4.5 years, and secondary market trackers like Caplight cite a much lower estimated valuation near $285M.

AI companies and labs · United States

Hyperscience

hyperscience.com ↗

Software that reads and processes documents for government agencies and big regulated companies.

Losing its protection

Assessed 21 July 2026 · Evidence as of 21 July 2026

The structural protection is eroding.

On the recordSince 21 July 2026 · Checkable from within 12 months · Every call on the record Checkpoint: Next funding round or recap, which will show if investors still back the 2021 valuation

The call

The government document business still runs, but the company hasn't raised money in over four years and keeps shrinking while it argues frontier AI needs a human-workflow layer on top.

Hyperscience is a software company that helps government agencies and banks automatically read and organize documents like tax forms and contracts. It has real advantages in this work because it meets strict security rules that Google, Microsoft, and Amazon's tools cannot meet, and this protection is hard to copy. But three problems are hitting hard at once: cheaper AI models are making basic document reading feel like a commodity that any company can offer, the company has not raised money in 4.5 years while other AI startups raise constantly, and it has cut its staff from 347 people to about 215 people and never fully bounced back. The biggest thing to watch is whether the company raises new money in the next year, which would signal the market still believes in its government-focused strategy, or whether it settles for a cheaper funding round that would suggest the business is under serious stress.

Why

AI pressure: GPT-4o, Gemini, and Claude doing zero-shot document extraction. Frontier models now read documents out of the box, pushing Hyperscience to sell itself as an orchestration layer instead of an extractor. Could AI do the core job? Only partly.

Bigger players: Microsoft Azure Document Intelligence, Google Document AI, AWS Textract. Hyperscalers ship competing document AI but stay locked out of FedRAMP High government work Hyperscience already holds. Could a giant copy it? Probably.

What protects it. FedRAMP High certification and entrenched SSA and VA workflows that took years to build and are hard to rip out.

What would change this call

Hyperscience announces a new funding round or valuation event within 12 months. An up-round confirms market confidence; a down-round or recap confirms the 4.5-year funding gap reflects distress.
Loss or non-renewal of a marquee government contract (SSA, VA, or DISA) becomes public. Would show the regulated-government moat weakening under budget or competitive pressure.
Microsoft, Google, or AWS announce a FedRAMP High authorization for their document AI products. Would show hyperscalers breaching the compliance barrier that currently protects Hyperscience's niche.

Key numbers

37% revenue growth in FY2024 (2025-05)
Headcount down 16% YoY to 215 employees (2026-06)
No funding round since December 2021's $1.65B valuation (2026-07)

Named risks

No new funding round since December 2021, over 4.5 years, with secondary signals implying steep valuation markdown. Severity: high
Headcount down 16% YoY to about 215 employees as of June 2026, never recovered from 2022 layoff. Severity: high
Frontier multimodal models commoditizing basic document extraction, eroding low-complexity use cases. Severity: medium
Heavy dependency on Palantir FedStart infrastructure and Carahsoft distribution for government contracts. Severity: medium
Broader IDP moat outside regulated government niche is thin against numerous co-Leader competitors. Severity: medium

Rivals, as read in July 2026

Rival	Direction	Evidence
UiPath (Document Understanding)	contesting	Co-Leader with Hyperscience in Gartner IDP Magic Quadrant, September 3, 2025
Frontier multimodal models (GPT-4o, Gemini, Claude)	losing to	Commoditizing basic zero-shot document extraction as of February 19, 2026
Microsoft Azure Document Intelligence, Google Document AI, AWS Textract	contesting	Hyperscalers ship competing IDP but remain blocked from FedRAMP High regulated workloads
Nanonets, Docsumo, Hypatos, Lido	gaining on	Self-serve, lower-cost SMB challengers undercutting enterprise pricing as of June 30, 2026
ABBYY Vantage, Rossum, Tungsten Automation	contesting	All named co-Leaders alongside Hyperscience in Gartner IDP MQ, September 3, 2025

Questions people ask

Is Hyperscience profitable?: Hyperscience does not disclose profitability. It reports 37% year over year revenue growth for FY2024 (ended Feb 28, 2025) with 90%+ gross margin, per company statements from May 5, 2026, but no net income or profit figures are published since it is a private company.
Has Hyperscience had layoffs?: Yes. In March 2022 Hyperscience cut about 100 employees, roughly 25% of staff, coinciding with CEO Peter Brodsky's departure. Headcount has not recovered since, standing at about 215 employees in June 2026, down 16% year over year from a peak of 347 in mid 2021.
Could AI replace Hyperscience?: Frontier multimodal models like GPT-4o, Gemini, and Claude can already perform zero-shot document extraction, commoditizing basic IDP use cases as of February 2026. Hyperscience is repositioning around FedRAMP High government compliance and an orchestration layer (Inference Layering Optimization, April 2026) rather than raw extraction to stay differentiated.
Who owns Hyperscience?: Hyperscience is privately held and venture backed, with no parent company. Investors include Bessemer Venture Partners, Battery Ventures, FirstMark, Stripes, and Tiger Global, as of a June 17, 2025 partnership announcement. Its last disclosed funding round was a $100M Series E in December 2021 valuing it around $1.65 billion.
Who is Hyperscience's biggest competitor?: Gartner's inaugural Magic Quadrant for Intelligent Document Processing, published September 3, 2025, named Hyperscience a Leader alongside ABBYY, Rossum, Tungsten Automation, and UiPath's Document Understanding module as co-Leaders, though none has publicly taken Hyperscience's government contract share.
Has Hyperscience raised new funding recently?: No. Hyperscience's last disclosed funding round was a $100M Series E in December 2021 at approximately $1.65 billion valuation. As of July 21, 2026, no new round has been reported in over 4.5 years, and secondary market trackers like Caplight cite a much lower estimated valuation near $285M.

Timeline

2026-06-25Won IDP Platform of the Year at AI Breakthrough Awards, second consecutive year
2026-05Named Leader and Customer Favorite in Forrester Wave: Document Mining and Analytics Platforms Q2 2026
2026-04-07Launched Spring 2026 Release with Inference Layering Optimization across CPUs, GPUs, and LLMs
2025-10-15Winter 2025 Release introduced ORCA vision-language model for zero-shot extraction
2025-09-03Named a Leader in Gartner's inaugural Magic Quadrant for Intelligent Document Processing
2025-06-17Partnered with Carahsoft as Master Government Aggregator for public sector distribution
2025-05-22Named launch partner for IBM watsonx Orchestrate's Agent Catalog
2024-12-17Achieved FedRAMP High authorization via Palantir's FedStart program
2024-06Launched Hypercell for GenAI, converting documents into LLM and RAG ready data
2022-03-03Laid off about 100 employees, roughly 25% of staff, CEO Peter Brodsky departed

Revision history

21 July 2026

First assessed. Losing its protection. The text above is fixed as of this date and is not edited afterwards.

In the same band

Safe Superintelligence Inc.Mistral AI Cohere