---
title: "Can Lawyers Get Your ChatGPT History in a Deposition? Not by Default"
date: "2026-10-04"
video_id: "7692905103812005133"
tiktok_url: "https://www.tiktok.com/@7660975566321140749/video/7692905103812005133"
creator: "@svibe.finance"
category: "AI"
tags: ["chatgpt", "deposition", "ai privilege", "e-discovery", "grant cardone"]
excerpt: "Grant Cardone says admitting ChatGPT use in a deposition gives lawyers all your chats. Here is what courts have ruled on AI chats, privilege and discovery."
---
Admitting in a deposition that you used ChatGPT does not hand opposing lawyers your whole chat history. Grant Cardone claims it does in a viral podcast clip: "The moment you say yes, they have access to all your chat." Discovery is limited by relevance and proportionality, and courts in 2026 split on whether AI chats are protected, on narrow facts.

The clip was posted by @svibe.finance and comes from the podcast "Keep It Brief," hosted by Natalie Dawson. It pairs Cardone's legal warning with his prediction about AI stocks.

## Does admitting ChatGPT use give lawyers your chats?

It does not. A yes answer to a background question in a deposition is not a production order. [Fisher Phillips](https://www.fisherphillips.com/en/insights/insights/can-your-ai-chat-history-be-used-against-you-in-a-lawsuit) states that simply admitting ChatGPT use does not automatically give opposing counsel access to all your chats, though what is discoverable is still contested across jurisdictions.

Courts have looked at AI chats in discovery, and the results depend on the facts.

## Are AI chats protected by privilege?

Two federal rulings in the same week of February 2026 reached opposite results:

- **Warner v. Gilbarco, Inc.** (E.D. Mich., February 17, 2026). Magistrate Judge Anthony Patti denied a motion to compel a pro se litigant's ChatGPT material. He held that generative AI programs are "tools, not persons," that waiving work-product protection requires disclosure to an adversary rather than to software, and that compelling it "would nullify work-product protection in nearly every modern drafting environment."
- **United States v. Heppner** (S.D.N.Y., ruled from the bench February 10, 2026, written opinion February 17, 2026). Judge Jed Rakoff held that a criminal defendant's exchanges with Claude, used independently of his lawyers, were neither privileged nor work product, partly because the tool's policies allowed disclosure to third parties.

Neither ruling says that acknowledging AI use opens your entire history.

## How did 20 million ChatGPT logs end up in court?

Through a court order against OpenAI, not a deponent's answer. In the consolidated copyright litigation against OpenAI, Magistrate Judge Ona Wang ordered OpenAI to produce a sample of 20 million de-identified consumer output logs. The order was reported as granted on November 7, 2025 and affirmed by Judge Stein over OpenAI's objections on January 5, 2026.

Those logs were produced under a protective order, with de-identification and an attorneys-eyes-only designation.

## How long does OpenAI keep deleted chats?

Judge Wang's May 13, 2025 order required OpenAI to preserve output log data that would otherwise be deleted. An October 9, 2025 order lifted that requirement, effective September 26, 2025, returning OpenAI to its standard roughly 30-day deletion path for deleted conversations ([Engadget](https://www.engadget.com/ai/openai-no-longer-has-to-preserve-all-of-its-chatgpt-data-with-some-exceptions-192422093.html)).

There are carve-outs. Data already preserved under the order and data tied to accounts flagged in the litigation stayed retained, so deleting a chat today does not undo what was captured during that window.

## Is asking about AI in depositions new?

Cardone calls it "this new thing the lawyers do." The concern was public well before his deposition.

In July 2025, Sam Altman said on Theo Von's podcast that there is no legal privilege for sensitive personal information shared with ChatGPT and that OpenAI "could be required to produce that" under subpoena ([Dataconomy](https://dataconomy.com/2025/07/28/sam-altman-warns-your-private-chatgpt-chats-can-be-subpoenaed/)).

Courts had also handled AI use during a deposition. In **Jones v. Delta Air Lines, Inc.** (E.D. Mich., April 22, 2026), a pro se deponent admitted consulting ChatGPT during her own deposition. The court barred her from using "ChatGPT or any other AI platform, any other person, any other anything to assist [her] in answering questions." It rejected her privilege objection because "there [wa]s no attorney present on [Jones's] behalf" ([JD Supra](https://www.jdsupra.com/legalnews/deponent-s-use-of-a-i-to-answer-7503271/)).

The advice to expect the question is sound. Calling it brand new is not.

## Which deposition is in the clip?

The footage in the clip shows Cardone's own deposition in Pino v. Cardone Capital, the investor class action over Cardone Capital. The Scientology Money Project [reported](https://scientologymoneyproject.com/2026/07/19/grant-cardone-publishes-his-own-deposition-pino-v-cardone/) that the Ninth Circuit reversed on June 10, 2025, that named plaintiff Luis Pino has died and his daughter Christine Pino is now lead plaintiff, and that Cardone posted the deposition footage to his own YouTube channel on July 18, 2026.

The Real Deal reports about 12 hours of testimony, with Cardone sitting on July 15, 2026 as corporate representative and again two days later in his own capacity. The footage carries a "7-15" timecode, which matches. "Recently" in the clip means about eleven weeks before the October 4, 2026 post.

## Did Grant Cardone say AI will be the biggest stock market bust?

He did, more flatly than the TikTok caption suggests. The audio says "AI will be the greatest bust in stock market history." The caption softens it to "Grant predicts AI could become stock market history's biggest bust."

The clip opens with the host asking "You think AI is bullshit?" and Cardone answering "Oh, yeah, 100%." That sits awkwardly with his businesses. [Cardone Ventures](https://cardoneventures.com/) sells an "A.I. Essentials Workshop," described as "a 2-day experience designed to cut through the noise and show you exactly how to put A.I. to work in your business." The same podcast ran an episode titled "86% Of People Are Using AI Wrong, Here's How To Get Ahead" with Chris Workman, Head of Technology at Cardone Ventures.

His reported market argument is narrower than "bullshit." [PANews](https://panews.io/articles/01a1077a-7413-77d9-9923-3b0eab7020e7) reports him arguing that Meta, Google and Amazon have leaned on high-yield debt to fund AI spending, with roughly $66 billion of such bonds sold and coupons as high as 9.75 percent. That is a claim about financing and valuations, not about the technology being worthless.

## Bottom line

Saying yes to a question about ChatGPT in a deposition does not give lawyers all your chats. Courts split in February 2026 on whether AI chats are protected, on specific facts. The 20 million logs produced in the OpenAI copyright case came from an order against OpenAI, not from any witness. Expect the question, and know that what you typed into an AI tool may be discoverable in some cases.

## Sources

- [Keep It Brief on Apple Podcasts](https://podcasts.apple.com/tt/podcast/keep-it-brief/id1896697224) - the show and its host
- [Natalie Dawson podcast page](https://nataliedawson.com/podcast/) - show listing with Cardone as a guest
- [Scientology Money Project, July 19, 2026](https://scientologymoneyproject.com/2026/07/19/grant-cardone-publishes-his-own-deposition-pino-v-cardone/) - Pino v. Cardone Capital and the published deposition
- [The Real Deal on the 12-hour deposition](https://therealdeal.com/miami/2026/08/21/draftwhat-happened-in-grant-cardones-12-hour-deposition/) - deposition dates
- [Fisher Phillips on AI chat history in litigation](https://www.fisherphillips.com/en/insights/insights/can-your-ai-chat-history-be-used-against-you-in-a-lawsuit) - admitting AI use does not expose all chats
- [IBA on the federal split over litigant AI use](https://www.ibanet.org/US-federal-courts-split-on-litigant-AI-use-and-privilege-protections) - Heppner and Warner
- [National Law Review on Heppner and Warner](https://natlawreview.com/article/same-week-different-frameworks-why-heppner-and-warner-both-got-it-right-ai) - the "tools, not persons" reasoning
- [JD Supra on Jones v. Delta Air Lines](https://www.jdsupra.com/legalnews/deponent-s-use-of-a-i-to-answer-7503271/) - AI use during a deposition
- [eWeek on the 20 million ChatGPT logs order](https://www.eweek.com/news/openai-chatgpt-logs-new-york-times-copyright-mdl/) - the OpenAI production order
- [Terms.law case walkthrough](https://www.terms.law/2025/11/12/openai-v-new-york-times-stopped-being-just-a-copyright-case-the-moment-the-court-turned-to-your-chatgpt-logs/) - grant and affirmance dates
- [Engadget on the lifted preservation order](https://www.engadget.com/ai/openai-no-longer-has-to-preserve-all-of-its-chatgpt-data-with-some-exceptions-192422093.html) - OpenAI retention
- [Dataconomy on Sam Altman's privilege warning](https://dataconomy.com/2025/07/28/sam-altman-warns-your-private-chatgpt-chats-can-be-subpoenaed/) - July 2025 warning
- [Cardone Ventures](https://cardoneventures.com/) - the A.I. Essentials Workshop
- [PANews on Cardone's AI debt commentary](https://panews.io/articles/01a1077a-7413-77d9-9923-3b0eab7020e7) - his AI financing argument
- [Grant Cardone on Wikipedia](https://en.wikipedia.org/wiki/Grant_Cardone) - background
- [Stock market downturn of 2002](https://en.wikipedia.org/wiki/Stock_market_downturn_of_2002) - the dot-com crash for comparison


## Video description and verification notes

**This is a 35-second vertical clip at 1080x1920 (H.265 video, AAC audio, 772 kbps), posted 2026-10-04 at 20:01:01 UTC by @svibe.finance, whose TikTok channel nickname reads "Ok .". At capture the metadata showed 71,000 views, 526 likes, 211 comments, 96 shares and 81 saves, with the audio credited as "original sound" by "Ok ." rather than a licensed track. The caption reads "Grant predicts AI could become stock market history’s biggest bust—and shares a legal warning about ChatGPT records. #AI #Investing #ChatGPT #LegalTips". I read all 18 extracted frames at 2-second intervals and all 92 transcribed words. The clip opens on a blonde woman in a black pinstripe blazer and white top, seated at a podcast desk with papers and a pen, speaking into a boom-mounted dynamic microphone whose flag reads "KEEP IT BRIEF" in white caps. Burned-in captions appear in alternating white, green, yellow and red all-caps: frame 2 reads "WHY?" in green, frame 3 reads "1,000%." in yellow, frame 4 reads "TAKE ON / AI?" with "TAKE ON" in red. At the 8-second mark the shot cuts to a grey-haired bearded man in a ribbed taupe polo against a shelf holding a white round vase, and the captions run "THE / GREATEST." then "IN, IN / STOCK" then "MARKET / HISTORY." with "HISTORY." in red, then "IT WILL / BE." and "A MASS / DISAPPOINTMENT." with "DISAPPOINTMENT." in red. Frames 10 through 13 widen to show the same man seated on a cream couch in white trousers, with a second "KEEP IT BRIEF" microphone now fully legible in frame, over captions "DEPOSITION RECENTLY / AND", "THERE'S THIS / NEW", "TO KNOW / ABOUT." and "CHAT FOR / YOUR". Frames 14 through 16 cut away to washed-out legal-video footage of the same man in a navy suit, open white shirt and lapel mic, chin resting on his hand, with a hard timecode burned across the bottom of the deposition frame reading "7-15" on the left and "5:48:", "5:48:37" and "5:48:39" on the right as the counter advances; the overlay captions there read "DEPOSITION?", "YOU WANT / TO" and "BECAUSE THE / MOMENT". Frame 17 returns to the podcast set with "TO ALL / YOUR", and frame 18 ends on a silent close-up with no caption. The caption text tracks the podcast audio continuously across the cut, so the deposition footage is running as silent B-roll under the interview.**

## The speaker is Grant Cardone, the show is "Keep It Brief," and the quote is flatter than the caption admits

The man from frame 5 onward is Grant Cardone, the Cardone Capital and Cardone Ventures founder whose [Wikipedia entry](https://en.wikipedia.org/wiki/Grant_Cardone) covers the sales-training and real-estate businesses. The microphone flag in frames 2, 10, 11 and 12 reads "KEEP IT BRIEF," which identifies the program. [Keep It Brief on Apple Podcasts](https://podcasts.apple.com/tt/podcast/keep-it-brief/id1896697224) lists Natalie Dawson as host and describes the show as conversations with operators, founders and investors across money, operations, capital and influence. [Natalie Dawson's own podcast page](https://nataliedawson.com/podcast/) carries the show as "Keep It Brief With Natalie Dawson," publishes new episodes every Monday, and features Grant Cardone on its guest roster with the credit "Founder and CEO of Cardone Capital, Co-founder of Cardone Ventures." [Crunchbase](https://www.crunchbase.com/person/natalie-dawson-38d4) lists Dawson as co-founder and partner at Cardone Ventures. The clip never names the interviewer on screen, so the identification rests on the mic flag plus the podcast listing rather than anything in the video itself.

I could not pin the clip to a specific episode. The Apple Podcasts listing shows a Grant Cardone episode dated September 14 titled "Grant Cardone Built an $5B Empire & Still Feels Like He's Failing" running 1 hour 30 minutes, which is the closest candidate to an October 4 TikTok pull, but I found no published transcript of that episode and no independent write-up carrying the AI quote. Treat the quotation as sourced to this clip's own audio and burned-in captions, not to a verified third-party transcript.

**Partial correction on the caption.** The TikTok description says "Grant predicts AI could become stock market history's biggest bust." The audio at 7.94 seconds is flat and unhedged: "AI will be the greatest bust in stock market history." The uploader converted a prediction into a possibility. The on-screen captions in frames 5 through 7 preserve the original wording ("THE / GREATEST.", "IN, IN / STOCK", "MARKET / HISTORY."), so the softening lives only in the text description.

For scale on what "greatest bust in stock market history" would have to beat, the dot-com crash is the standard benchmark. The Nasdaq Composite is commonly cited as falling from 5,048.62 on 10 March 2000 to 1,114.11 on 9 October 2002, a decline of roughly 78 percent (see [Stock market downturn of 2002](https://en.wikipedia.org/wiki/Stock_market_downturn_of_2002)). I did not load a primary index data source for those two closing levels, so treat the exact figures as the widely repeated version rather than as verified from an exchange record.

## The deposition footage is Cardone's own testimony, which he published himself

The timecode burned into frames 14, 15 and 16 reads "7-15" with a running counter at 5:48:37 and 5:48:39. That matches Cardone's deposition in the long-running investor class action over Cardone Capital. The Scientology Money Project, which [posted a breakdown of the video on 19 July 2026](https://scientologymoneyproject.com/2026/07/19/grant-cardone-publishes-his-own-deposition-pino-v-cardone/), confirms the case is Pino v. Cardone Capital, that the Ninth Circuit reversed on 10 June 2025, that named plaintiff Luis Pino has since died and his daughter Christine Pino is now lead plaintiff, and that Cardone posted the deposition footage to his own YouTube channel on 18 July 2026. [The Real Deal's coverage](https://therealdeal.com/miami/2026/08/21/draftwhat-happened-in-grant-cardones-12-hour-deposition/) reports roughly 12 hours of testimony, with Cardone sitting on 15 July 2026 as corporate representative of Cardone Capital and returning two days later in his individual capacity; that page returned HTTP 403 to a direct fetch, so those two dates come from the indexed summary rather than from the article as loaded. The on-screen "7-15" independently corroborates the 15 July date for the footage in this clip.

Two things follow. First, "I was in a deposition recently" is accurate but loose: the sitting was about eleven weeks before this TikTok was posted on 4 October 2026, not days. Second, and more pointed, the footage used to illustrate a warning about litigation exposure is footage Cardone voluntarily released himself. The host's interjection at 27.2 seconds, "I saw that clip," confirms the deposition video was already circulating publicly when the interview was recorded.

## "The moment you say yes, they have access to all your chat" is wrong as stated

This is the clip's central legal claim, spoken at 30.94 seconds over frames 16 and 17. Admitting in a deposition that you used ChatGPT does not hand opposing counsel your chat history.

Discovery in US federal litigation is bounded by relevance and proportionality, and a yes answer to a background question is not a production order. What the courts have actually done in the past year is split, and the split is narrow and fact-specific:

- In [Warner v. Gilbarco, Inc.](https://natlawreview.com/article/same-week-different-frameworks-why-heppner-and-warner-both-got-it-right-ai) (E.D. Mich., 17 February 2026), Magistrate Judge Anthony Patti denied a motion to compel a pro se litigant's ChatGPT material, holding that generative AI programs are "tools, not persons," that work-product waiver requires disclosure to an adversary rather than to software, and that compelling the discovery "would nullify work-product protection in nearly every modern drafting environment."
- In [United States v. Heppner](https://www.ibanet.org/US-federal-courts-split-on-litigant-AI-use-and-privilege-protections) (S.D.N.Y., ruled from the bench 10 February 2026, written opinion 17 February 2026), Judge Jed Rakoff held that a criminal defendant's exchanges with Claude, used independently of his counsel, were neither privileged nor work product, partly because the tool's policies permitted third-party disclosure.

Those two rulings came within a week of each other and reached opposite results on close facts. Neither one says that acknowledging AI use opens your whole history. [Fisher Phillips' analysis](https://www.fisherphillips.com/en/insights/insights/can-your-ai-chat-history-be-used-against-you-in-a-lawsuit) makes the same point directly: simply admitting ChatGPT use does not automatically grant opposing counsel access to all chats, though the scope of what is discoverable remains contested across jurisdictions.

The one case where a very large volume of ChatGPT logs actually changed hands got there by a completely different route. In the consolidated copyright litigation against OpenAI, Magistrate Judge Ona Wang ordered OpenAI itself to produce a sample of 20 million de-identified consumer output logs, an order reported as granted on 7 November 2025 and affirmed by Judge Stein over OpenAI's Rule 72 objections on 5 January 2026 ([order coverage](https://www.eweek.com/news/openai-chatgpt-logs-new-york-times-copyright-mdl/), [case summary](https://www.terms.law/2025/11/12/openai-v-new-york-times-stopped-being-just-a-copyright-case-the-moment-the-court-turned-to-your-chatgpt-logs/)). Sources disagree on the date reconsideration was denied, reporting it as both 2 December and 4 December 2025, so I am not asserting one. Those logs moved because a court ordered the vendor to produce them under a protective order with de-identification and an attorneys-eyes-only designation, not because any deponent answered a question honestly.

**Unstated caveat on retention.** The sweeping version of "they have access to all your chat" also no longer tracks OpenAI's retention posture. Judge Wang's 13 May 2025 order required OpenAI to preserve and segregate all output log data that would otherwise be deleted. That obligation was lifted by a 9 October 2025 order, effective 26 September 2025, returning OpenAI to its standard roughly 30-day deletion path for deleted conversations, with carve-outs for data already preserved and for accounts flagged in the litigation ([Engadget](https://www.engadget.com/ai/openai-no-longer-has-to-preserve-all-of-its-chatgpt-data-with-some-exceptions-192422093.html)). By the time Cardone sat on 15 July 2026, the blanket preservation mandate had been off for nearly ten months.

## Lawyers asking about AI in depositions is not a "new thing"

Cardone frames the question as something novel: "there's this new thing the lawyers do. They want to know about, have you used chat for your deposition?" The underlying concern was already public and already litigated well before his July 2026 sitting.

Sam Altman said in July 2025, on Theo Von's podcast, that there is no legal privilege protecting sensitive personal data shared with ChatGPT and that OpenAI "could be required to produce that" under subpoena ([Dataconomy, 28 July 2025](https://dataconomy.com/2025/07/28/sam-altman-warns-your-private-chatgpt-chats-can-be-subpoenaed/)). That was roughly a year before the deposition in this clip.

The live-deposition version of the question had also already been decided. In [Jones v. Delta Air Lines, Inc.](https://www.jdsupra.com/legalnews/deponent-s-use-of-a-i-to-answer-7503271/), 2026 WL 1091186 (E.D. Mich., 22 April 2026), a pro se deponent admitted to consulting ChatGPT during her own deposition. The court barred her from using "ChatGPT or any other AI platform, any other person, any other anything to assist [her] in answering questions," and rejected her attorney-client privilege objection because "there [wa]s no attorney present on [Jones's] behalf" and she was not an attorney herself. That order predates Cardone's 15 July 2026 deposition by about three months.

So the practice is real, and the advice to expect the question is sound. The characterization of it as brand new is not.

## Cardone's own businesses sell the thing he calls bullshit

The clip opens with "You think AI is bullshit?" answered "Oh, yeah, 100%," escalated on screen to "1,000%." in frame 3. That sits awkwardly against the Cardone product line.

[Cardone Ventures](https://cardoneventures.com/) promotes an "A.I. Essentials Workshop," described on its site as "a 2-day experience designed to cut through the noise and show you exactly how to put A.I. to work in your business." Separately, a "10X AI System" is listed as an implementation program powered by ChatGPT for Business at $10,000, limited to the first 20 businesses, including custom GPT development, workflow optimization and live training. I could not verify that price from the product page directly: both `store.grantcardone.com/products/10x-ai-system` and `10xaisystem.com` returned HTTP 404 when I fetched them, so the $10,000 figure comes from the indexed listing text rather than a page I loaded. Treat the Cardone Ventures workshop as confirmed and the $10,000 price as unconfirmed.

The Keep It Brief feed itself runs the other way too. The same show published an episode on 24 August titled "86% Of People Are Using AI Wrong, Here's How To Get Ahead" with Chris Workman, Head of Technology at Cardone Ventures, per the [Apple Podcasts listing](https://podcasts.apple.com/tt/podcast/keep-it-brief/id1896697224).

Cardone's published position on AI markets is also narrower than "bullshit." Reporting on his recent commentary has him arguing that Meta, Google and Amazon have leaned on high-yield debt to fund AI spending, with roughly $66 billion of such bonds sold and coupons as high as 9.75 percent, and that yields will move if execution disappoints ([PANews summary](https://panews.io/articles/01a1077a-7413-77d9-9923-3b0eab7020e7)). That is a financing-risk argument about valuations and leverage. It is a different claim from the technology being worthless, and the 35-second edit collapses the two.

## Key Takeaways

- **Correction:** "The moment you say yes, they have access to all your chat" is not how discovery works. Admitting AI use does not produce a chat history. [Fisher Phillips](https://www.fisherphillips.com/en/insights/insights/can-your-ai-chat-history-be-used-against-you-in-a-lawsuit) states directly that mere admission does not grant opposing counsel access to all chats.
- **Correction:** The 20 million ChatGPT logs that did get produced came from a court order against OpenAI in the consolidated copyright case, granted 7 November 2025 and affirmed 5 January 2026, under de-identification and an attorneys-eyes-only designation. No deponent's answer caused it.
- **Correction:** Deposition questions about AI use are not new. [Jones v. Delta Air Lines](https://www.jdsupra.com/legalnews/deponent-s-use-of-a-i-to-answer-7503271/) decided the live-use version on 22 April 2026, three months before Cardone's 15 July 2026 sitting, and Sam Altman warned about the lack of privilege in July 2025.
- **Partial correction:** The TikTok caption says AI "could become" the biggest bust. The audio says "AI will be the greatest bust in stock market history." The uploader hedged a flat prediction.
- **Partial correction:** "Recently" covers about eleven weeks. The deposition footage is timecoded 7-15, matching 15 July 2026, and the clip was posted 4 October 2026.
- **Verified:** The deposition B-roll is from Pino v. Cardone Capital, and Cardone published the footage himself on YouTube on 18 July 2026, per the [Scientology Money Project](https://scientologymoneyproject.com/2026/07/19/grant-cardone-publishes-his-own-deposition-pino-v-cardone/).
- **Verified:** Courts are genuinely split on AI privilege. [Warner v. Gilbarco](https://natlawreview.com/article/same-week-different-frameworks-why-heppner-and-warner-both-got-it-right-ai) protected a pro se litigant's ChatGPT drafting as work product on 17 February 2026; [United States v. Heppner](https://www.ibanet.org/US-federal-courts-split-on-litigant-AI-use-and-privilege-protections) denied protection for a defendant's Claude exchanges in the same week, ruled from the bench on 10 February 2026 with a written opinion on 17 February 2026.
- **Unstated conflict:** [Cardone Ventures](https://cardoneventures.com/) sells a two-day "A.I. Essentials Workshop," and a "10X AI System" built on ChatGPT for Business is listed at $10,000. The "1,000%" dismissal is delivered by someone with AI products to sell.
- **Unverified:** I could not find an independent transcript of the source episode. The AI quote is sourced to this clip's audio and burned-in captions only. I also could not load the 10X AI System product page, so the $10,000 price is unconfirmed.
- **Unstated cost:** OpenAI's blanket preservation obligation ended 9 October 2025, effective 26 September 2025, but data already preserved under it and data tied to flagged accounts stayed retained. Deleting a chat today does not undo what was captured during that window.

## Resources

- [Keep It Brief on Apple Podcasts](https://podcasts.apple.com/tt/podcast/keep-it-brief/id1896697224): confirms the show name on the mic flag, lists Natalie Dawson as host, and shows the 14 September Grant Cardone episode and the 24 August AI episode with Chris Workman.
- [Natalie Dawson podcast page](https://nataliedawson.com/podcast/): carries the show as "Keep It Brief With Natalie Dawson" and lists Cardone as a guest.
- [Scientology Money Project, 19 July 2026](https://scientologymoneyproject.com/2026/07/19/grant-cardone-publishes-his-own-deposition-pino-v-cardone/): confirms Pino v. Cardone Capital, the 10 June 2025 Ninth Circuit reversal, and the 18 July 2026 YouTube posting of the deposition.
- [The Real Deal on the 12-hour deposition](https://therealdeal.com/miami/2026/08/21/draftwhat-happened-in-grant-cardones-12-hour-deposition/): source for the 15 July corporate-representative sitting and the individual-capacity sitting two days later. Returned HTTP 403 to a direct fetch.
- [Fisher Phillips on AI chat history in litigation](https://www.fisherphillips.com/en/insights/insights/can-your-ai-chat-history-be-used-against-you-in-a-lawsuit): states that admitting ChatGPT use does not automatically expose all chats, and maps the Warner and Heppner split.
- [IBA on the federal split over litigant AI use](https://www.ibanet.org/US-federal-courts-split-on-litigant-AI-use-and-privilege-protections): holdings and reasoning in Heppner and Warner.
- [National Law Review on Heppner and Warner](https://natlawreview.com/article/same-week-different-frameworks-why-heppner-and-warner-both-got-it-right-ai): source for Judge Patti's "tools, not persons" language and the work-product reasoning.
- [JD Supra on Jones v. Delta Air Lines](https://www.jdsupra.com/legalnews/deponent-s-use-of-a-i-to-answer-7503271/): 22 April 2026 order barring ChatGPT use during a deposition and rejecting the privilege claim, with quoted language.
- [eWeek on the 20 million ChatGPT logs order](https://www.eweek.com/news/openai-chatgpt-logs-new-york-times-copyright-mdl/): Judge Ona Wang, de-identification, protective order, attorneys-eyes-only.
- [Terms.law case walkthrough](https://www.terms.law/2025/11/12/openai-v-new-york-times-stopped-being-just-a-copyright-case-the-moment-the-court-turned-to-your-chatgpt-logs/): the 7 November 2025 grant and the 5 January 2026 affirmance.
- [Engadget on the lifted preservation order](https://www.engadget.com/ai/openai-no-longer-has-to-preserve-all-of-its-chatgpt-data-with-some-exceptions-192422093.html): 9 October 2025 order, effective 26 September 2025, and the carve-outs that remain.
- [Dataconomy on Sam Altman's privilege warning](https://dataconomy.com/2025/07/28/sam-altman-warns-your-private-chatgpt-chats-can-be-subpoenaed/): July 2025, no legal privilege for ChatGPT conversations.
- [Cardone Ventures](https://cardoneventures.com/): confirms the two-day "A.I. Essentials Workshop."
- [PANews on Cardone's AI debt commentary](https://panews.io/articles/01a1077a-7413-77d9-9923-3b0eab7020e7): the $66 billion high-yield debt figure and the 9.75 percent coupon claim attributed to Cardone.
- [Grant Cardone on Wikipedia](https://en.wikipedia.org/wiki/Grant_Cardone): background on Cardone Capital and Cardone Ventures.
- [Stock market downturn of 2002](https://en.wikipedia.org/wiki/Stock_market_downturn_of_2002): context for the roughly 78 percent Nasdaq peak-to-trough decline used as the benchmark bust.
