---Advertisement---

Parag Agrawal, Ex CEO,Twitter launched his new AI starts up “Parallel Web Systems”.

Published On: August 18, 2025
Follow Us
Parag Agarwal launches his AI startup.
---Advertisement---

Parag Agrawal, former CEO of Twitter,fired by Elon Musk in 2022, launched AI startup, Parallel Web Systems. This AI startup systems is for large-scale online research. Established in 2023, the company has secured good amount of funding from investors like Khosla Ventures, First Round Capital and Index Ventures.

The Open Web’s Next Chapter: Why AI Demands a New Internet. Why Parag Agrawal and his team starting up new AI venture ?

The open web is one of humanity’s most transformative inventions—a dynamic, decentralized space where knowledge flows freely, fueling innovation, collaboration, and discovery. It’s the foundation upon which modern AI was built. But now, the very system that enabled AI’s rise is facing an existential challenge: the web’s primary users are no longer just humans—they’re AIs.

The Coming AI Web Revolution

Today, human interaction with the web is simple: we click, scroll, and search, consuming information in bite-sized chunks. AI, however, interacts with the web in entirely different ways—sometimes needing a single fact in milliseconds, other times ingesting entire databases for hours of analysis.

The problem? The internet wasn’t built for this.

Current web infrastructure—designed around human attention—relies on ads, paywalls, and fragmented APIs. As AI becomes the dominant user, these models break down, pushing the web toward a fractured future of walled gardens, proprietary data silos, and restricted access. If we don’t act, the open ecosystem that birthed AI could collapse under its own success.

Building a Programmatic Web for AI

To avoid this, we need a fundamental redesign—a Programmatic Web, engineered for AI’s needs:

  • Unified Data & Computation – A seamless fusion of knowledge, reasoning, and processing, where outputs aren’t just documents but actionable insights.
  • Declarative Interfaces – Instead of AIs digging through layers of links, they should simply declare what they need—let the infrastructure handle the how.
  • Transparent Attribution – Every piece of data, every insight, must be traceable to its source, ensuring fairness, trust, and accountability.
  • Open, Value-Driven Markets – Contributors should be rewarded based on the value they provide, making openness not just idealistic, but economically inevitable.

Why This Matters

This isn’t about nostalgia for the old web—it’s about unlocking the next era of innovation. A Programmatic Web means:

  • AI solving problems at unprecedented scale—from medical research to climate modeling.
  • Accelerating discovery by breaking down knowledge barriers.
  • Enabling entirely new forms of creativity and automation we can’t yet imagine.

The choice is stark: Either we rebuild the web for AI, or we watch it splinter into irrelevance. The future belongs to those who design for abundance, not scarcity—and that future starts now.

The “Parallel Web System” linkedin acount narrates ….

“The Parallel Task API now supports Auto Mode, enabling one-off web research queries without requiring explicit output schemas.
Simply ask a question. Our processors will then conduct research and generate a structured output schema for you.
AIs will soon use the internet far more than humans ever have. At Parallel, we are building for the web’s second user.Our API is the first to surpass humans and all leading AI models (including GPT-5) on deep web research tasks.

Parallel’s performance on independent benchmarks:

  • BrowseComp: 58% accuracy vs GPT-5 (41%) vs humans (25%)
  • DeepResearch Bench: 82% win rate over reference vs GPT-5 (66%). Parallel beats GPT-5 head-to-head 74% of the time.

For $0.10, we exceed what takes humans 2 hours on BrowseComp.

The open web’s greatest chapter is just beginning.

One of his team member Todd Jackson posted on Linkedin that

Todd Jackson, team member of Parallel Web System. Photo: linkedin

“Immensely proud to be able to partner with Parag and the entire Parallel team. They’ve made incredible progress in a short amount of time and I can’t wait for more folks to start playing around with the APIs they’ve built.”

His another team member Travers Nisbet posted on linkedin that thay starte Parallel Web System because the primary user of the web is changing: from humans to AIs.

Travers Nisbet , member of Parallel Web System. Photo : linkedin

The internet’s next chapter won’t be written by humans alone. As AI becomes the dominant user, we must rebuild the web’s foundations to serve this new era.

At Parallel, they are not just adapting to this shift—also accelerating it. Because when AI can navigate the web as fluidly as humans—only smarter and faster— unlock possibilities human only begun to imagine.

*Source : linkedin.com

S ROY

Tech enthusiast and MSc. in Electronics. MTech. in Environmental Management. Part time research scholar (2012-2019) on various topics at University of Kalyani. West Bengal,India. Interested in Artificial Intelligence, Neuroscience, Environmental technology,Human health- tech cordination. Human Mental Health, Biophysics. Diploma in Yoga and Naturopathy fron West Bengal Council of Yoga and Naturopathy. NDDY from Gandhi Smarak Prakitik Chikitsalaya. Teaching Experience at Govt Primary School. More than 19 years of experience of Operator post job at Thermal Power Plant. Research journals on SARS-CoV2 and Waste Water treatment and management. Now Independent Researcher. Works as an author for https://www.techbyte365.com.

Join WhatsApp

Join Now

Join Telegram

Join Now

Leave a Comment