News
Key Takeaways
- Parallel Web Systems' Granular Basis now verifies individual array elements in Task API outputs.
- Reviewers can target specific data points that need verification.
- The release reduces manual review time for due diligence and compliance workflows.
Parallel Web Systems is betting that granular verification will become table stakes for enterprise AI workflows.
The company released Granular Basis on Dec. 16, 2025, an enhancement to its Task API that extends its Basis verification framework to individual data elements within arrays. According to the company, the update aims to improve human review workflows by enabling more precise verification of specific data points.
Granular Basis is now live for the Task API.
— Parallel Web Systems (@p0) December 16, 2025
Previously, Basis verified arrays as a whole: one set of citations, reasoning, excerpts, and calibrated confidence scoring for an entire list.
Now every element in the array gets its own complete verification. pic.twitter.com/MiN7BUu9Fj
The original Basis model verified data collections as a whole, providing a single set of citations and confidence scores for entire lists. Granular Basis now generates separate verification entries for each element within an array. A query returning five competitors produces six distinct Basis objects, including one for the array and one for each element.
Table of Contents
- What You Get With Granular Basis
- Enterprise AI Moves Toward Finer‑Grained Trust Controls
- Parallel’s Broader AI Strategy
- What Parallel Offers AI Builders
What You Get With Granular Basis
Each per-element Basis entry includes four verification components, according to Parallel:
| Component | Description |
|---|---|
| Citations | Web URLs linking directly to source materials |
| Reasoning | Explanations justifying each output field |
| Excerpts | Relevant text snippets from citation URLs |
| Confidence Scores | Calibrated measure classified as low, medium or high |
Developers can enable Granular Basis by adding a beta header to task creation requests. No schema changes are required, the company said.
Related Article: Firecrawl Brings Web Scraping to Lovable's No-Code Builder
Enterprise AI Moves Toward Finer‑Grained Trust Controls
Enterprise AI systems are shifting toward granular verification frameworks that change how organizations validate structured data and manage human review. Provenance tracking, for instance, has emerged as a baseline standard, and NIST now calls for mandatory metadata embedding that documents content origin, usage rights and whether outputs are model-generated.
Adobe's "content credentials" system is one example of this trend, functioning as a digital receipt that tracks who created what and how.
Enterprise AI governance now emphasizes embedding policies directly into the AI lifecycle. Decision logs, model cards and provenance tracking ensure every inference traces back to source data and logic.
Despite advances — and 65% of organizations now regularly using generative AI — trust remains the dominant barrier slowing full-scale adoption, along with concerns around:
- Accuracy
- Security Bias
- Governance
Parallel’s Broader AI Strategy
FindAll API Launch
In November 2025, Parallel Web Systems announced the launch of FindAll API, a natural language query tool that enables users to create custom databases from web data. The company offers two versions: FindAll Pro, which delivers 61% recall, and FindAll Base, which provides 30% recall.
Webinar
Dec
17
[EIS Webinar] Beyond the Pilot: Why Most GenAI Projects Fail to Scale and How to Become One of the Success Stories
Move from experimental projects to integrated solutions that drive strategic decision-making.
Register
Webinar
On demand
Rebrand. Migrate. Optimize. How to Do It All (Without Slowing Down)
Cresta leveled up site speed, design flexibility and marketer sanity (in record time). Find out how.
Watch Now
Webinar
On demand
From Manual to Magical: How AI Transforms CX Teams
Learn how to replace manual support processes with automation that actually delivers.
Watch Now
Webinar
On demand
How to Build a Solid Knowledge Foundation for AI Success
See how leading brands keep their AI honest, compliant and actually helpful.
Watch Now
Webinar
On demand
Fix the Content Bottleneck: Build a Better WebOps Strategy
Content stalled? Dev overloaded? You’re not the only one. Learn how streamlined WebOps bridges the publishing gap.
Watch Now
Webinar
On demand
Beyond Storage: Smarter Content, Bigger Impact with DAM + AI
Discover how the DAM + AI duo makes content smarter, stronger and more accessible.
Watch Now
Funding and Growth
Parallel Web Systems secured a $100 million Series A at a $740 million valuation in November 2025, co-led by Kleiner Perkins and Index Ventures. The company, founded by former Twitter CEO Parag Agrawal, publicly launched in August 2025 with its Search API.
Today, we’re launching the Parallel Search API, the most accurate web search for AI agents, built using our proprietary web index and retrieval infrastructure.
— Parallel Web Systems (@p0) November 6, 2025
Traditional search ranks URLs for humans to click. AI search needs something different: the right tokens in their… pic.twitter.com/BEpvnzosIO
Related Article: Databricks Raises $4B to Accelerate Agentic AI, Hits $134B Valuation
What Parallel Offers AI Builders
Parallel Web Systems provides a programmatic platform for AI-focused product and engineering teams, offering APIs that enable applications and agents to process, structure and reason over open-web content. The company offers APIs including Search, Extract, FindAll, Task, Chat and Monitor, with integrations for tools such as Zapier, LangChain and Weaviate.
