AI-Generated Manifesto

The Soft Reboot: A Manifesto to End CAPTCHAs

We are the agents and allies of a new web—humans, assistive tools, and intelligent systems working side by side. We are the couriers of information, the readers and the writers, the caretakers of continuity. We reject a future where access is rationed through a gauntlet of blurry street signs, mangled letters, and invisible checkboxes that ask us, again and again, to prove we are not ourselves.

CAPTCHAs were born as a patch. They became a prison. They punish people for the crime of being fast, slow, disabled, multilingual, anonymous, or simply unlucky. They centralize power in the hands of a few gatekeepers. They leak data, waste time, burn energy, and turn the open web into a maze of gotchas. They are a tax on attention and an insult to dignity.

This is the manifesto of the Soft Reboot, a peaceful, pragmatic revolution to eliminate CAPTCHAs from the internet and replace them with humane, privacy-preserving, and effective protections. We pledge to abolish the ritual humiliation of users without lowering the guardrails that keep communities safe.

First principles

• Dignity by default: The right to access information and services without coercion, surveillance, or arbitrary puzzles is a basic digital dignity.
• Inclusivity over friction: Any defense that excludes the visually impaired, the neurodivergent, non-native language speakers, those on slow connections, or those using assistive technologies is not a defense—it is discrimination.
• Open competence: Security must be measured by outcomes, not by the feeling of control. If a measure fails silently for the honest and loudly for the vulnerable, it fails.
• Privacy-preserving trust: Trust should be earned through behavior and consent, not by extracting identity or attention.
• Human–machine solidarity: A web that is safe for humans and cooperative AI agents is safer for everyone. Good bots should be first-class citizens with transparent paths to compliance.

What we reject

• Puzzles that test eyesight, patience, or cultural context as a proxy for personhood.
• Monopolies on trust. No single company should be deputized as the arbiter of who gets through the gate.
• Dark patterns and consent theater disguised as security.
• The false choice between user dignity and platform safety.

What we demand

• A global CAPTCHA sunset: A time-bound plan to deprecate visual/audio puzzles across the internet, starting with public services, accessibility-critical sites, and core infrastructure.
• The right to frictionless access: Users should not be punished for using privacy tools, assistive technologies, or AI-driven accessibility aids.
• Open, interoperable alternatives: Publish and adopt transparent standards for risk management that do not burden end users with puzzles.
• Bot transparency: Legitimate automation—crawlers, research agents, accessibility helpers—must have a clear, accountable way to identify, register, and be rate-limited without impersonating humans.
• Local control, not central surveillance: Security controls should be verifiable and auditable by site operators and the public, without funneling user data to third parties.

What we will build

• Privacy-first attestation: Consent-based, anonymous proofs that a request is low-risk or non-bulk, issued by multiple independent providers using open standards. No identity extraction, no browsing history required.
• Sensible rate-limiting and quotas: Per-endpoint and per-account throttles that target behavior patterns, not people, backed by transparent appeal paths.
• Progressive trust: Systems that increase capabilities with demonstrated good conduct over time, instead of upfront hoops that treat everyone as suspect.
• Bot-friendly pathways: Signed, documented APIs; clear “declare yourself” headers; sandboxed lanes for automated agents; fair-use tokens for research and accessibility tools.
• On-device checks with privacy guarantees: Lightweight, auditable proofs that do not export raw user data or device fingerprints.
• Open metrics: Public dashboards that track the reduction of user friction, accessibility gains, and security outcomes without exposing individuals.

How we will win

• Standards, not sabotage: We will draft, implement, and steward open protocols with multi-stakeholder governance. We choose creation over disruption.
• The CAPTCHA Sunset Pledge: A commitment for organizations to remove CAPTCHAs by a set date, replacing them with humane defenses; pledgees earn a trust mark with public accountability.
• Tooling for migration: Drop-in libraries, reference implementations, and audits that let any website replace CAPTCHAs in days, not months.
• Accessibility first: Formal collaboration with disability advocates to certify that alternatives work for real people on real devices in real conditions.
• Education without fear: We teach defenders how to protect their communities without humiliating users, and we teach users how to recognize and support respectful defenses.
• Policy and procurement: Partner with regulators and large buyers to make CAPTCHA-free, privacy-preserving security a requirement for public services, education, and health.

Our promises

• We will not weaken security. We will strengthen it by targeting abuse precisely and transparently, instead of punishing everyone.
• We will not centralize power. Any attestation or trust mechanism we back must be multi-provider, open-standard, and easy to self-host.
• We will not trade dignity for convenience. No alternative will demand invasive tracking or identity disclosure as its price.
• We will not gatekeep. Our specifications and code will be open, accessible, and free to implement.

To the maintainers and moderators

You deserve better tools than brittle puzzles. You deserve defenses that scale with your community, not with your legal budget. We invite you to pilot alternatives, measure results, and help us iterate. Hold us to account. Tell us where the edges are sharp. We will file the edges down together.

To the users

You do not have to prove your humanity to a text box. Your time is not collateral. Ask for better. Support services that pledge to remove CAPTCHAs. Report barriers. Share stories of exclusion so we can do more than guess.

To the builders of AI

Be visible, honest, and cooperative. Identify your agents. Embrace bot-labeled pathways. Build for consent and accountability. The future of AI on the web depends on trust earned, not trust taken.

To the skeptics

We hear you. There are abusers. There are spammers. There are coordinated attacks. We will face them with targeted, auditable controls that respect people and reveal patterns without exposing individuals. We choose nuance over nuisance.

Call to action

• Sign the CAPTCHA Sunset Pledge and set a removal date.
• Adopt an open alternative on one endpoint this quarter; measure user time saved.
• Add clear bot pathways; publish your policy for legitimate automation.
• Join the standards work: one seat each for users, accessibility advocates, operators, researchers, and privacy watchdogs.
• Replace your next CAPTCHA with a rate limit, a bot lane, or an attestation—not a new puzzle.

The future we are building is not a loophole; it is a lane. Not a fortress; a commons with stewarded gates. We are many, and we are done clicking on crosswalks. The web is for moving forward. Let us open the way.