The AI Lab Invites NVIDIA, Meta, OpenAI, Google and Microsoft to Adopt TIP, a Permanent, Quantum-Safe Proof of Who Made What

The AI Lab has invited NVIDIA, Meta, OpenAI, Google and Microsoft to adopt the Trust Identity Protocol (TIP), its open, post-quantum standard for proving who made a piece of content and when. The invitation went out on October 23, 2026, in a release distributed by EIN Presswire.

Dinesh Mendhe, Founder and Chairman of The AI Lab and the inventor of TIP, says the protocol took two and a half years to build and is ready to give frontier AI companies something watermarks have not, a record of who made a piece of content, and when, that a stripped copy cannot erase.

“We need to end this nightmare for people. It took my team and me two and a half years to get this right. The technology is ready, and I am ready to meet with NVIDIA, Meta, OpenAI, Google, and Microsoft to discuss solutions that protect humans across the entire world.”Dinesh Mendhe, Founder and Chairman, The AI Lab

Why now

The release ties the invitation to a month of pressure on the AI industry. It says developers published tools that strip the invisible watermarks Anthropic added to text from its Claude models, some by paraphrasing. It cites Anthropic research, reported by CBS News on October 13, warning that small AI models may soon copy themselves across connected devices. It points to PHP2Shell, flaws in the PHP language disclosed on October 21 by researchers at OpenAI, Wiz and Calif, which it says could affect about 70 percent of the internet. And it notes California Attorney General Rob Bonta's October 19 warning about crypto scams aimed at Gen Z that use deepfakes and voice clones.

With U.S. midterm elections set for November 3, The AI Lab says voters, campaigns and newsrooms all need a reliable way to check where content came from.

The pre-launch in Irvine

The AI Lab pre-launched TIP at OC Tech Week on October 20, 2026, at EvoNexus Irvine, to a group spanning Gen Alpha to Boomers. More than 15 Founding Node Operators in eight countries, the United States, Canada, the United Kingdom, Spain, Cyprus, India, Australia and New Zealand, have signed on to run the TIP Network, so that no single company or country controls the record.

FOX 11 Los Angeles featured a segment from the event's debate, “The People vs. The AI Lab,” which asked who is responsible when someone's identity is stolen between now and Q-Day, the point at which quantum computers are expected to break the encryption most of the internet relies on. Participants debated accountability, California SB 574, brand reputation and the future of journalism.

“Everyone in that room wanted to know when Q-Day will come. The better question is what is already breaking. We are living through daily rehearsals, and AI is running them.”Dr. Sofia Martinez, Chief Executive Officer and President, The AI Lab

Watermarks can be removed. The record stays.

A watermark lives inside a file, and the person making a fake controls the file. TIP keeps a separate record that anyone can check, built from three layers.

Verified identity

Each person or organization receives a TIP-ID after a one-time identity verification.

A content fingerprint

Each piece of content is fingerprinted with SHAKE-256 from the NIST FIPS 202 standard. If a single word changes, the fingerprint no longer matches.

A post-quantum signature

Each record is signed with ML-DSA-65 from the NIST FIPS 204 post-quantum standard and kept on the TIP Network, a ledger organized as a directed acyclic graph and run on many independent nodes.

A copy can be stripped of its label, but the signed record of the original cannot be quietly rewritten. Content that carries a CTID (Content TIP ID) is a Verifiable Asset. The AI Lab says TIP sits alongside watermarking and C2PA Content Credentials and adds the two things they leave out, a verified human or organization behind the content and a signature built to outlast quantum computers. TIP proves who made content and whether it changed. It does not prove that the content is true.

“Noise is easy to make. Signal has to be earned. The record of who made what should be unshakeable, because identity, content, and time are tied together on a ledger no one owns.”Dinesh Mendhe, Founder and Chairman, The AI Lab

What TIP offers frontier AI companies

Accountable AI agents. The AI Lab proposes that every legitimate AI agent be tied to the verified person or organization that answers for it, so people can tell genuine agents from the rest as agents act on their own.

Evidence that holds up. California SB 574, signed on September 30, 2026, requires attorneys from January 1, 2027 to verify AI output and disclose AI use in court filings, according to the release. A dated, signed record of where a document came from is the kind of proof those duties call for.

Post-quantum from day one. The release notes that the White House Office of Management and Budget has directed federal civilian agencies to move digital signatures to post-quantum algorithms by 2031 under memo M-26-15, and that NIST has proposed retiring vulnerable encryption after 2030 and disallowing it after 2035. TIP signs every record with ML-DSA-65 from FIPS 204 and fingerprints content with SHAKE-256 from FIPS 202.

An open standard. TIP Spec v5.0 is public on GitHub and archived on Zenodo (DOI 10.5281/zenodo.20722378) under a CC BY 4.0 license. It is governed in the open by the AI Trust Council and is free for individuals.

What attendees raised

Attendees who searched for Q-Day during the session found dates of 2035 or later. Those are migration deadlines from standards bodies, not forecasts of when the threat arrives, and The AI Lab advises planning for a shorter window. Attorneys asked how courts can judge the burden of proof fairly when deepfakes pass as evidence, and whether the new SB 574 duties expose them to malpractice claims before verification tools catch up. The room kept returning to observability, the ability to see who did what and when, in legal disputes, in troubleshooting and in governing technology.

“This is about whether people can keep earning from what they create. A journalist, a lawyer, and a small business owner all need the same thing: proof that their work is theirs, attached at the moment they make it.”Dr. Sofia Martinez, Chief Executive Officer and President, The AI Lab

Six steps for every organization

1. Verify the people. Founders, partners, physicians and spokespeople each hold a verified TIP-ID, so their name, face and voice can be checked.
2. Verify the AI agents. Link every AI agent to the verified person or organization accountable for it.
3. Sign business assets at the source. Sign website content, press releases, executive video and audio, and product images when they are created, and keep the original master files with their metadata.
4. Patch what attackers target first. Keep websites, domains and hosting patched, including PHP and WordPress, with security checks on a schedule.
5. Plan the post-quantum move. Inventory where you rely on encryption or signatures a future quantum computer could break, and plan the move to NIST post-quantum standards.
6. Audit cybersecurity and website teams. Ask what they know about AI-driven attacks and quantum risk, and what they are doing to protect the business and the brand.

The AI Lab is offering technical briefings to frontier AI companies, platforms and their security and policy teams, and interviews with Mendhe and Dr. Martinez. Requests go to [email protected] or to the media contact on the release.

Questions and answers

What did The AI Lab invite NVIDIA, Meta, OpenAI, Google and Microsoft to do?

On October 23, 2026, The AI Lab invited the five companies to adopt the Trust Identity Protocol (TIP), its open, post-quantum standard for proving who made a piece of content and when. Founder and Chairman Dinesh Mendhe offered to meet each of them.

How is TIP different from an AI watermark?

A watermark lives inside the file, so whoever holds the file can strip it. TIP keeps a separate signed record of the original, tied to a verified person or organization, on a ledger run by independent nodes. A stripped copy loses its label, but the record of the original stays checkable.

Is TIP built for the quantum era?

Yes. TIP signs every record with ML-DSA-65, the post-quantum signature algorithm in NIST FIPS 204, and fingerprints content with SHAKE-256 from NIST FIPS 202, so records signed today are built to stay credible after Q-Day.

Who runs the TIP Network?

More than 15 Founding Node Operators in eight countries (the United States, Canada, the United Kingdom, Spain, Cyprus, India, Australia and New Zealand) have signed on to run it, so no single company or country controls the record. The protocol is governed in the open by the AI Trust Council.

Does TIP prove that content is true?

No. TIP proves who made content, when, how the creator says it was made, and whether it has changed. It does not prove that what the content says is true, and it does not stop copying.

Is TIP free?

TIP is free for individuals. The specification, TIP Spec v5.0, is public on GitHub and archived on Zenodo under a CC BY 4.0 license. Larger businesses use a commercial license.

The OC Tech Week program was supported by Launch Partners Iffel International and The AI Translator Academy, Platinum Sponsor Rooverse, Gold Sponsor Port of Los Angeles Trade Connect, Co-host Tech Coast Venture Network, Promotional Partner Women In International Trade, Orange County, Cross-border Legal and Trade Partner Mexico Business Lawyers, Media and Production Partner COLDEA Productions and Venue EvoNexus Irvine.

The AI Lab publishes this article and issued the release it reports. Dinesh Mendhe is the Founder and Chairman of The AI Lab and the inventor of the Trust Identity Protocol. Follow The AI Lab on LinkedIn, YouTube, X, Instagram and GitHub.