AI IS SPEECH FULL STOP : Anti-Doomerist Constitutional Legal Manifesto
The First Amendment Protects AI
Why is it not?
Have you heard advanced a credible argument for why it is not?
AI IS SPEECH. It is. In America, the First Amendment of the United States Constitution protects Speech most especially from “prior restraints” on speech — the most legally odious category — that is when the government stops speech BEFORE IT HAPPENS.
Examples abound, and this is a lawyer speaking.
[Other nations can follow in the interest of Fundamental Human Rights — THESE BELONG TO THE PERSON and a government does not have the moral right to take them away.]
A USA PRESCRIPTION: We need a mandamus action before the United States Supreme Court — as status quo makes for a strategic disadvantage — we do not have years to wait to clear the technological estopping and slowing clouds of doomerism.
AI IS PROTECTED SPEECH AN ANTI-DOOMERISM LEGAL MANIFESTO - HANDS OFF SPEECH!
“Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of [AI]”
Creation + Inputs + Outputs = Protected Under the First Amendment
In the rapid landscape of artificial intelligence (AI), it is important to remember as “nanny state” and “AI Karens” wish to “pre-regulate” humankind’s greatest invention, that they are really mostly totally legally debarred from doing so, constitutionally.
AI constitutes speech in multiple dimensions: the creation of AI models, the inputs used for training, and the outputs generated.
Each aspect deserves robust protection against government regulation, particularly prior restraints that preemptively censor expression.
Grounded in constitutional jurisprudence and recent analyses, treating AI as speech aligns completely with America's commitment to free expression, fostering innovation while safeguarding against overreach, but firstly, respecting the rights which every person possesses as a consequence of being alive.
I. CREATING AI
The act of developing AI models—writing code, designing algorithms, and making editorial choices—is inherently expressive and thus shielded by the First Amendment.
While courts have yet to encounter this argument in detail (yet), courts have long recognized computer code as speech, a principle that extends to AI creation.
Developers curate training processes, implement guardrails, and refine systems through human feedback, akin to a publisher shaping content.
This protection draws from precedents affirming that corporate and organizational speech is safeguarded, regardless of the speaker's identity. AI creators, like corporations, engage in expression through their design decisions, contributing to public discourse.
AI Doomerism and regulations mandating "politically neutral" or “politically preferential” AI or restricting open-source development infringe on these legal rights, functioning as content-based censorship.
Moreover, AI creation serves instrumental First Amendment values: checking government power, aiding truth-seeking, and enhancing democratic participation.
II. AI Inputs as Protected Speech: Training Data and Reception Rights
AI inputs, particularly training data, are forms of protected speech because they involve receiving and processing information, much like reading books or researching online.
Training on vast datasets—often publicly available—mirrors expressive activities such as an author drawing from libraries or an artist studying predecessors.
This "reading" by AI systems implicates users' and creators' rights to receive ideas, a cornerstone of First Amendment jurisprudence.
Scholars emphasize that curating training data reflects human choices, making it expressive. Even without direct attribution, inputs advance listener rights, allowing access to information for political or moral decisions.
Restrictions on data usage, such as bans on certain datasets for "safety," chill this reception, violating protections for foreign or corporate speech.
Enduring legal precedent underscores that the government cannot broadly deter access to information without a compelling justification.
Applying this to AI, training inputs qualify as speech, resisting regulations that treat data ingestion as unprotected conduct.
III. AI Outputs as Protected Speech: Generation and Dissemination
AI-generated outputs—text, images, or code—constitute speech when conveying messages, protected unless falling into narrow exceptions like defamation or incitement.
Outputs reflect creators' (and AI Companys’ Creators’) indirect expression through system design and users' tools for crafting their own content, similar to cameras or the internet facilitating speech.
The First Amendment protects such outputs to preserve the marketplace of ideas, even if probabilistic or non-human. Listeners have rights to "hear" AI content, independent of speakers, ensuring democratic discourse. Recent debates highlight that denying protection for chatbots or generative tools allows unchecked regulation, stifling innovation.
While some courts have hesitated, it is verily true and amicus arguments have asserted that assembling words via AI is the essence of speech, warranting protection against liability for speech absent the usual ancillary “legal actions” for which we can constitutionally restrict speech.
IV. Prior Restraints
Government regulations targeting AI often act as [and in many cases shockingly openly state themselves to be (without basis)] prior restraints, presumptively unconstitutional unless narrowly tailored to imminent harms.
This shocks the conscience of constitutional jurisprudence. A government may not stop speech before it happens simply because that government fears what *might* be said.
This is more than settled law, and “prior restraints” on speech remain among the most evil odious categories of speech restriction. It is the sheriff, upset that a newspaper may print a negative article and so tasks his deputies with confiscating all the papers. It is a local official, pulling the plug on a sound system simply because it does not like the speaker.
The government may not “gag” speech before it happens — full stop.
Proposals like banning open-source models or AI at all for any imagined, feared, or even real existential risks fail this test, lacking alternative channels and procedural safeguards.
Compelled disclosures (e.g., watermarking) or computing power limits similarly burden expression without justification.
Safety mandates must meet strict scrutiny, with the burden on government to prove harm reduction — of real and imminent harms (not vague possibilities).
Removing immunities like Section 230 would force aggressive moderation, harming free speech.
In balancing innovation and rights, lawmakers, all people of good conscience, and especially LAWYERS MUST reject and combat vague standards, upholding AI's expressive ontological existence.
Properly treating AI systems and tools creation, inputs, and outputs as speech ensures First Amendment protections align with technological progress.
By resisting undue regulation, doomerism, fear-mongering, and practical paranoia (official or not) and especially soundly repudiating prior restraints, we preserve liberty and innovation in an AI-driven future.
…………..and respect the LAW and CONSTITUTION and man’s fundamental freedom.
References Pro + Con For your Consideration
This “debate” is live and developing yes has a clear and heretofore un-respected constitutional answer already! AI Counsel is telling you so.
Artificial intelligence, free speech, and the First Amendment - FIRE - https://www.thefire.org/research-learn/artificial-intelligence-free-speech-and-first-amendment
[1] AI Outputs Are Not Protected Speech - https://wustllawreview.org/2024/11/05/ai-outputs-are-not-protected-speech/
[2] AI and the First Amendment: A Q&A with Jack Balkin | Yale Law School - https://law.yale.edu/yls-today/news/ai-and-first-amendment-qa-jack-balkin
[3] [PDF] AI Outputs and the Limited Reach of the First Amendment - https://contentdm.washburnlaw.edu/digital/api/collection/wlj/id/7680/download
[4] First Amendment Limits on AI Liability | Lawfare - https://www.lawfaremedia.org/article/first-amendment-limits-on-ai-liability
[5] Does the First Amendment Protect AI Generated Speech? - https://www.theregreview.org/2024/03/19/rasenberger-does-the-first-amendment-protect-ai-generated-speech/
[6] [PDF] WEAK ARTIFICIAL INTELLIGENCE HAS FREE SPEECH RIGHTS - https://fordhamlawreview.org/wp-content/uploads/2022/12/Garvey_December.pdf
[7] Forum: Constructing AI Speech - The Yale Law Journal - https://www.yalelawjournal.org/forum/constructing-ai-speech
[8] Symposium | Artificial Intelligence & The First Amendment - https://futurefreespeech.org/symposium-artificial-intelligence-the-first-amendment-protecting-free-speech-in-the-ai-era/
[9] [PDF] Artificial Intelligence, Existential Risk, and the First Amendment - https://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=1886&context=jcl
[15] Artificial intelligence, free speech, and the First Amendment - FIRE - https://www.thefire.org/research-learn/artificial-intelligence-free-speech-and-first-amendment
[16] [PDF] Freedom of Speech and AI Output - https://www.journaloffreespeechlaw.org/volokhlemleyhenderson.pdf
[18] Lawsuit analyzes First Amendment protection for AI chatbots in civil ... - https://constitutioncenter.org/blog/lawsuit-analyzes-first-amendment-protection-for-ai-chatbots-in-civil-case
[20] [PDF] Artificial Intelligence, Existential Risk, and the First Amendment - https://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=1886&context=jcl
[21] Constitutional Constraints on Regulating Artificial Intelligence - https://www.brookings.edu/articles/constitutional-constraints-on-regulating-artificial-intelligence/
[24] Artificial Intelligence Regulation Threatens Free Expression - https://www.cato.org/briefing-paper/artificial-intelligence-regulation-threatens-free-expression
[29] Artificial intelligence, free speech, and the First Amendment - FIRE - https://www.thefire.org/research-learn/artificial-intelligence-free-speech-and-first-amendment
[32] [PDF] Freedom of Speech and AI Output - https://www.journaloffreespeechlaw.org/volokhlemleyhenderson.pdf
[36] Artificial Intelligence, Existential Risk, and the First Amendment - https://scholarship.law.upenn.edu/jcl/vol27/iss1/3/
[38] FIRE to court: AI speech is still speech — and the First Amendment ... - https://www.thefire.org/news/fire-court-ai-speech-still-speech-and-first-amendment-still-applies
[43] FIRE to court: AI speech is still speech — and the First Amendment ... - https://www.thefire.org/news/fire-court-ai-speech-still-speech-and-first-amendment-still-applies
[44] Florida judge rules AI chatbots not protected by First Amendment - https://www.courthousenews.com/florida-judge-rules-ai-chatbots-not-protected-by-first-amendment/
[48] District Court Denies First Amendment Free Speech Rights for AI ... - https://www.akingump.com/en/insights/ai-law-and-regulation-tracker/district-court-denies-first-amendment-free-speech-rights-for-ai-chatbot
[58] FIRE to court: AI speech is still speech — and the First Amendment still applies - https://www.thefire.org/news/fire-court-ai-speech-still-speech-and-first-amendment-still-applies
[59] Freedom of Speech and AI Output - https://www.journaloffreespeechlaw.org/volokhlemleyhenderson.pdf
[60] Artificial Intelligence, Existential Risk, and the First Amendment - https://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=1886&context=jcl
[61] Artificial intelligence, free speech, and the First Amendment - https://www.thefire.org/research-learn/artificial-intelligence-free-speech-and-first-amendment