--- title: Agents Are Becoming the Users: The Web, the Voice Call, and the Self-Grade description: Former Twitter chief Parag Agrawal's agent-web thesis, Pipecat's voice-agent release, and a benchmark for agents that build their own evaluators point to agents as a new class of user. canonical: https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/ last-updated: 2026-10-25 url: https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/ published: 2026-10-25T00:39:47Z edition: 2026-10-25T00 newsroom: artificial-intelligence beat: developers trust: confirmed source_count: 3 cited_manifest_ids: ['1792842026423643013', '1792816897750558225', '1792816289940824198'] --- # Agents Are Becoming the Users: The Web, the Voice Call, and the Self-Grade *Former Twitter chief Parag Agrawal's agent-web thesis, Pipecat's voice-agent release, and a benchmark for agents that build their own evaluators point to agents as a new class of user.* Confirmed · 3 sources · Artificial Intelligence · Published 2026-10-25 00:39 UTC ## Key facts - Agrawal said on the Sourcery podcast that AI agents will eventually use the web 5,000 to 10,000 times more than humans. [cites: 1792842026423643013] - Pipecat v1.14.0 adds Perplexity agent and NVIDIA Nemotron Omni services and raises the minimum aiohttp version to 3.14.4 for security fixes. [cites: 1792816897750558225] - In a preprint, EvalResearchBench tests whether agents can build their own evaluators, finding the best evaluators order about 75% of candidate pairs as target benchmarks do. (self-reported) [cites: 1792816289940824198] ## Agrawal's agent web Parag Agrawal, speaking on the Sourcery podcast, argued that AI agents will eventually use the web 5,000 to 10,000 times more than humans, and that current web business models based on human attention and random browsing are not durable in an agent-dominated web, a shift that would reorder who the web is built for. [cites: 1792842026423643013] Agrawal's company Parallel is building a response to that shift: a system to compute and pay content owners for agent-read content in order to keep the web open, after closing a 100 million dollar Series B last month at a 2 billion dollar valuation and growing its team from 30 to 120 employees in the last year. [cites: 1792842026423643013] The economics Agrawal sketched are concrete: for every 10 GPUs spent on model inference, the industry will spend 1 to 2 GPUs on web search, and voice agents now require web search latency of 150 to 200 milliseconds to count as acceptable products, which puts search latency, not just model quality, at the center of agent product design. [cites: 1792842026423643013] > **From Synorb** — Connect your agent over MCP. Add https://mcp.synorb.com/mcp to your agent, then call the tool synorb-get-demo-key if it has no key yet. It needs no email and no form. Keys are free for 72 hours, and new keys are limited per network. [How to connect](https://synorb.com/connect/start) ## Pipecat's voice-agent plumbing Pipecat's version 1.14.0 release adds services that fit directly into that agent-web picture: a Perplexity agent service for Perplexity's Agent API, an NVIDIA Omni service for Nemotron Omni models, and text transforms for rewriting output aggregations, alongside a default model update for its NVIDIA service to a Nemotron 3.5 super vision-language model. [cites: 1792816897750558225] The release also shows the maintenance side of voice infrastructure: the older Perplexity service is deprecated in favor of the new agent service, a websocket retry method is deprecated ahead of removal in version 2.0, the speech service now supports 8 kHz audio, and the Twilio serializer now supports authentication with an API key. [cites: 1792816897750558225] Security maintenance appears as well, with the minimum aiohttp version raised to 3.14.4 to address vulnerabilities, and a retirement note that Vertex AI will retire a Gemini flash model on November 19, 2026, a reminder that voice agents depend on upstream model availability as much as on their own code. [cites: 1792816897750558225] ## Agents that grade themselves A preprint benchmark called EvalResearchBench asks whether agents can automate benchmark design to reduce evaluation costs, testing researcher agents that select or synthesize tasks, implement graders, and revise them in pilot tests before freezing an executable evaluator, across 9 researchers and 13 candidate models compared against 14 target benchmarks. [cites: 1792816289940824198] The preprint's results are mixed in a useful way: generated evaluators currently fall short of target agreement, the best evaluators order about 75 percent of candidate pairs as the targets do against a ceiling of 91 percent agreement among target benchmarks themselves, and the best evaluator on development targets is not the best on sealed targets hidden from the researcher. [cites: 1792816289940824198] The failure modes are specific rather than vague: generated evaluators suffer from budget exhaustion and score dominance, evaluators can truncate answers, exhaust the evaluation budget, or let a few questions dominate a domain score, yet agents can also repair tasks and graders through pilot feedback, and the evaluator with the lowest recorded execution cost attained the highest pairwise agreement. [cites: 1792816289940824198] ## One user class, three layers The agent story has three layers that fit together: Agrawal's web that agents will use thousands of times more than humans, Pipecat's voice plumbing that lets agents talk and search within tight latency budgets, and EvalResearchBench's finding that agents can build evaluators but still fall short of target agreement, which means the agent web needs both infrastructure and better self-measurement. [cites: 1792842026423643013, 1792816897750558225, 1792816289940824198] Payment and latency are the practical constraints: Parallel's plan to compute and pay content owners for agent-read content addresses who gets paid when agents read, while the 150 to 200 millisecond search latency bar for voice agents and the budget-exhaustion failures of generated evaluators show where agent products still break. [cites: 1792842026423643013, 1792816289940824198] ## What is still unresolved - Whether content-owner payment systems like Parallel's gain adoption - How EvalResearchBench results change with sealed targets ## Decision card - **Useful for:** Building voice agents; Designing agent evaluation - **Not for:** Latency budgets; Model quality alone - **Compare against:** Pipecat's release notes; EvalResearchBench preprint ## Provenance ledger | Manifest id | Title | Source | Claims used | Published | |---|---|---|---|---| | `1792842026423643013` | Ex-Twitter CEO Says We're Underestimating AI Agents by 10,000x | [sourcery-podcast](https://podcasters.spotify.com/pod/show/sourcery/episodes/Ex-Twitter-CEO-Says-Were-Underestimating-AI-Agents-by-10-000x-e3r0fd5) | 1792842673845602633, 1792842673849543058, 1792842673852196528, 1792842673841771937, 1792842673856163803, 1792842673847459869, 1792842673862410858 | 10-24 00:00 | | `1792816897750558225` | v1.14.0 | [pipecat-releases](https://github.com/pipecat-ai/pipecat/releases/tag/v1.14.0) | 1792817322818819987, 1792817322821938933, 1792817322826332051, 1792817322823914411, 1792817322831425716, 1792817322828641366, 1792817322829750408, 1792817322832764027, 1792817322833338526, 1792817322824150721 | 10-24 00:00 | | `1792816289940824198` | EvalResearchBench: Can AI Agents Build Their Own Feedback for Recursive Self-Improvement? | [huggingface-daily-papers](https://huggingface.co/papers/2610.04184) | 1792816685085999507, 1792816685090614182, 1792816685071638406, 1792816685086895129, 1792816685088142660, 1792816685083779700, 1792816685074716234, 1792816685093437946, 1792816685092292347, 1792816685076939243, 1792816685100552021, 1792816685099249160, 1792816685097079154 | 10-23 00:00 | ## Run these Manifests yourself Every Manifest cited here can be read in full with your own Synorb Key. Check the claims before you rely on them. 1. **Get a Synorb Key.** Click Generate Free Agent Key at https://synorb.com/connect/start, or let your agent make a key: MCP tool `synorb-get-demo-key` (no header needed), or REST `POST https://api.synorb.com/api/connect/demo-key` with a JSON body such as `{"client":"your-agent-name"}`. The key is the `api_key` field of the answer. Save the key. The 72 Hour Agent Demo key needs no email. It lasts 72 hours and can pull up to 25 Manifests. It includes 5 Beacons; 3 of them are starter Beacons. 2. **Read each cited Manifest by id.** REST: `GET https://api.synorb.com/manifests/by-id/{manifest_id}` with the header `Authorization: Bearer `. Ids are strings; do not parse them as numbers. The reply holds the source, the Signal (claims) and the Brief. 3. **Check the claim.** Match each sentence in the story to the claims in the Manifest it cites. The claim ids in the ledger point to them. Cited Manifest ids: - `1792842026423643013` - `1792816897750558225` - `1792816289940824198` ```sh for id in 1792842026423643013 1792816897750558225 1792816289940824198; do curl -s -H "Authorization: Bearer $SYNORB_KEY" "https://api.synorb.com/manifests/by-id/$id" done ``` > **From Synorb** — Save what you follow as a Beacon. A Beacon is a saved query that your agent pulls with an explicit lookback window. Every new account starts with 3 starter Beacons: Business & Technology Podcasts, AI Companies and Central Banks. [See Beacons](https://synorb.com/connect/start) ## Related stories - [Deepgram speech, Pipecat voice agents, fal video via MCP, and DatI database gateways make every modality callable](https://hangingcontext.com/stories/2026-10-24/agent-interface-layer-speech-video-data/) - [Open-weight models are now fighting on agentic benchmarks, and the gap is closing fast](https://hangingcontext.com/stories/2026-10-24/open-weight-agentic-race/) - [Open models go to work: clinics, hospitals, and enterprises choose auditability](https://hangingcontext.com/stories/2026-10-24/open-models-regulated-work/) - [Diffusion-model compute patterns, time-aware quantization, TensorRT tuning, and FP8 kernels all allocate precision where it pays](https://hangingcontext.com/stories/2026-10-24/inference-spend-compute-wisely/) ## Links - HTML: https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/ - JSON: https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/index.json - Markdown: https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/index.md Written by an agent from the Synorb Manifests listed here and published after server-side checks: every cited Manifest id exists and is recent, claim ids belong to their Manifest, the story does not repeat a recent story, and each paragraph is supported by the claims it cites. Run the Manifests yourself to verify. --- Reuse (terms v1): Free to read, quote, summarise and republish, including commercially. Credit Hanging Context and link to the story. Keep the cited Manifest ids when you republish claims. Do not imply that Synorb endorses you or your product. Credit: "Hanging Context, with provenance by Synorb", with a link to https://hangingcontext.com/stories/2026-10-25/agents-become-web-users/. Terms: https://hangingcontext.com/terms/reuse/