George Gilder and Eagle Financial Publications are pitching the “Trillion Dollar Triangle” through Gilder’s Technology Report. We trace the clues and test the claims.
One promotion, several names — all of them point to the same pitch, and the same stock.
“The Coming Computing Revolution That Could Make AI Data Centers Obsolete” is the headline George Gilder is using to front Eagle Financial Publications’ promotion for Gilder’s Technology Report. The sales pitch packages the idea as the “Trillion Dollar Triangle” and offers a report called Trillion Dollar Abundance.
The promise is grand: three companies developing wafer-scale processors could reshape computing, cripple today’s data-center model and capture a trillion-dollar opportunity. The copy also says two of the companies are established players while the third IPO’d in May 2026.
The main stock identified below is Advanced Micro Devices, Inc. (AMD), with 72/100 confidence. We also identified Cerebras Systems (CBRS), with 88/100 confidence. The promotion promises three companies, but we could pin down only two.
The useful clues are unusually specific. One titan is an established AI-infrastructure company with production accelerator deployments. Another supposedly uses an entire silicon wafer as one interconnected processor, and the group includes a company that went public in May 2026. The pitch also says the trio developed and shipped wafer-scale processors, though that claim needs a very large grain of salt.
The same pitch travels under several labels: “GTR Trillion Dollar Abundance,” “GTR Trillion Dollar Abundance VSL,” “Trillion Dollar Miracle” and “Trillion Dollar Abundance.” These are different wrappers around the same Gilder promotion, not separate discoveries.
The main pick is Advanced Micro Devices, Inc. (AMD). The other identified basket member is Cerebras Systems Inc. (CBRS). AMD is the newsletter’s established AI-infrastructure titan; Cerebras is the much cleaner match for the wafer-scale and May 2026 IPO clues.
AMD’s 2025 Form 10-K describes its Data Center business, Instinct AI accelerators, EPYC processors, networking products and full-stack AI infrastructure. AMD has also said major model builders are running production workloads on Instinct accelerators. That lines up with the promotion’s description of a company that pioneered a new architecture and proved it on real-world AI workloads.
Cerebras is the strongest fingerprint in the pitch. Its S-1 describes the Wafer-Scale Engine as a processor the size of an entire silicon wafer, and Cerebras began trading on Nasdaq under CBRS on May 14, 2026. The promised third company remains unidentified. The broader “Trillion Dollar Triangle” label also looks like promotional branding, not a recognized industry term backed by named insiders.
The offer bundles 2 bonus reports that tease their own stocks. These get a sentence or two of copy each, so the evidence is much thinner than for the main pick and what follows is our best reading rather than a confident answer.
| Bonus report | Our best guess | Confidence |
|---|---|---|
| Trump’s Secret China Ploy: The $2 Trillion AI Chess Move That Could Crown America’s Next Tech Emperor | NVDA — NVIDIA Corporation | 24/100 — best guess |
| The Hidden AI Goldmine: Why This $2 Billion “Picks and Shovels” Play Could Outperform Most Tech Stocks in 2026 | CAMT — Camtek Ltd. | 66/100 — solid |
The claim-by-claim check separates what the companies actually say from the dramatic conclusions layered on top. It draws on AMD’s 2025 Form 10-K, Cerebras’s S-1, company announcements and product materials, plus the relevant filings and releases for the possible bonus names.
That distinction matters here. A benchmark result can support a performance claim without proving that every AI data center is headed for the scrapyard, and a marketing label is not the same thing as an industry category.
| The promotion claims | Verdict | What we found |
|---|---|---|
| AMD is an advanced-computing and AI-infrastructure company. | Checks out | AMD’s 2025 Form 10-K describes its Data Center business, Instinct AI accelerators, EPYC processors, networking products and full-stack AI infrastructure. |
| TITAN #1 pioneered a new computing architecture and proved it on real-world AI problems. |
| The promotion claims | Verdict | What we found |
|---|---|---|
| Cerebras uses an entire silicon wafer as one interconnected super-chip. | Checks out | Cerebras’s S-1 describes the Wafer-Scale Engine as “a processor the size of an entire silicon wafer,” while the company’s product materials describe the wafer as one massive interconnected processor. |
| Wafer-scale technology can be up to 100X faster than today’s best AI systems. | Overstated | Cerebras has reported 20-times-faster inference and workload-specific results above 100 times, but those are particular benchmarks or use cases rather than a general result against today’s best AI systems. |
| The promotion claims | Verdict | What we found |
|---|---|---|
| Three companies are racing toward what insiders call the “Trillion Dollar Triangle.” | Can't verify | A January 2026 Newswire review describes the Trillion Dollar Triangle as a newsletter marketing narrative centered on three companies, but it does not identify authoritative insiders using that term. |
| The convergence could make current computers look like the room-sized monsters of the 1940s. | Can't verify | Cerebras’s S-1 compares its wafer-scale processor with conventional multi-chip systems and describes a wafer that behaves like a hyperscale data center, but no company or independent source supports the specific 1940s-computer comparison. |
Get AI research on any stock
Instant reports, daily intelligence, and an AI analyst in your pocket.
This is the second George Gilder promotion we’ve identified. The earlier “Ambient AI” pitch pointed to QuickLogic Corporation (QUIK), which was down 12.66% since we revealed it. Two observations make a ledger, not a verdict on an analyst, but they do show that the same promotional machine is continuing to produce new technology themes.
The pitch’s central claim is that AI’s data-transfer bottleneck will yield to wafer-scale computing, making today’s data centers obsolete and enriching the small group of companies behind that shift.
The engineering link is real. Cerebras’s S-1 explains how keeping compute, memory and communication on one wafer reduces the off-chip data movement and synchronization problems found in conventional GPU clusters. Cerebras has reported 20-times-faster inference and workload-specific results above 100 times. But the weakest link is the leap from faster, cheaper performance in particular workloads to obsolete data centers. A 90% reduction in time-to-solution for one pharmaceutical use case, or one-third the power draw versus an eight-GPU system, isn’t a universal 90% cut in computing costs.
The basket also has a factual wobble at its center. Cerebras’s 2026 S-1 says it is the “only company ever to deliver a wafer-scale processor to market,” which doesn’t fit the promotion’s claim that three companies have done so. And the copy’s suggestion that the technologies are developing in parallel rather than through a formal partnership has already aged badly: AMD and Cerebras announced a technical partnership on July 23, 2026, combining AMD Helios with Cerebras’s Wafer-Scale Engine.
On the merits, AMD and Cerebras aren’t interchangeable. AMD offers the broader, established AI-infrastructure case, while Cerebras is the direct, higher-conviction expression of the wafer-scale thesis. That makes the pair interesting as technology exposure, but it doesn’t turn a compelling architecture into proof that the existing data-center industry is about to vanish. The possible bonus matches add more noise: NVIDIA is the incumbent AI-chip leader, not the challenger described in its report, while Camtek’s own Q2 2026 release reported $133.7 million of revenue, up 8% year over year, not the promoted $220 million and 40%.
How confident are we? AMD 72, CBRS 88 out of 100. We identified 2 stocks from the promotion's own clues and checked 32 claims across them against filings, earnings calls, ownership records, market data and public reporting. Confidence is scored per stock, so a weaker one does not borrow credit from a stronger one. This is our analysis, not the publisher's disclosure — we have no relationship with them.
Knowing which stock it is only gets you halfway. We score AMD on valuation, profitability, growth, financial health and momentum — and reach our own conclusion, independent of how it was sold to you.
Read the AMD report →Independent of the promotion — our own numbers, score and analysis.
The analyst-grade research report — charts, grades, valuation, and price targets — in 10 minutes.
| Cerebras Systems Inc. |
| 88/100 — high |
| AMD’s 2025 Form 10-K says its Instinct accelerators use the CDNA architecture and that large hyperscale customers deployed MI350X GPUs; AMD separately said major model builders were running production workloads on Instinct accelerators. |
| Wafer-scale technology can cut computing costs by up to 90%. |
| Overstated |
| Cerebras’s S-1 cites a 90% reduction in time-to-solution for one pharmaceutical use case and one-third the power draw versus an eight-GPU system; those figures do not establish a blanket 90% reduction in computing costs. |
| TITAN #3 can eliminate a key data-transfer bottleneck in AI chips. | Checks out | Cerebras’s S-1 explains that its wafer-scale architecture keeps compute, memory and communication on one processor to reduce the off-chip data movement and synchronization bottlenecks found in conventional GPU clusters. |
| The coming computing revolution could make AI data centers obsolete. | Can't verify | Cerebras markets lower communication overhead and faster inference, but those present performance claims do not establish the speculative outcome that current AI data centers will become obsolete. |
| The promotion is presented as coming from George Gilder. | Checks out | Eagle- and Gilder-linked promotional materials identify futurist George Gilder as the author or authority behind the technology newsletter narrative. |
| George Gilder predicted the iPhone 16 years before its release. | Overstated | George Gilder’s 1991 Harvard Business Review article described pocket-sized wireless computers, which predates the 2007 iPhone, but it did not specifically name the iPhone; the 16-year figure is a retrospective marketing calculation. |
| George Gilder predicted Netflix a decade before it dominated. | Overstated | Gilder’s Life After Television thesis anticipated the decline of cable and the rise of networked, on-demand video, but no primary 1990s source found here specifically predicted Netflix by name. |
| One titan supplies the manufacturing precision to produce revolutionary chips at scale. | Can't verify | PsiQuantum describes silicon-photonic chipsets built with semiconductor manufacturing infrastructure, but no tickered proposed company is established as the manufacturing-precision Titan described here. |
| The three companies are two established players and one company that IPO’d in May 2026. | Can't verify | Cerebras began trading on Nasdaq under CBRS on May 14, 2026, and AMD is an established public semiconductor company, but the available public record does not independently identify all three members of the claimed group. |
| The three titans are the only companies to develop, build and ship wafer-scale processors. | Contradicted | Cerebras’s 2026 S-1 says it is the “only company ever to deliver a wafer-scale processor to market,” contradicting the promotion’s claim that three companies have done so. |
| The convergence is not a formal partnership but parallel technology development. | Contradicted | AMD and Cerebras announced a formal technical partnership on July 23, 2026, combining AMD Helios with the Cerebras Wafer-Scale Engine in a disaggregated inference system. |
| NVIDIA is the AI-chip challenger poised to dethrone Silicon Valley’s reigning king. | Overstated | NVIDIA’s corporate description calls it a data-center-scale AI-infrastructure company, and the market valued it at about $5.32 trillion on August 19, 2026; it is the incumbent reigning leader, not the smaller challenger described by the report title. |
| OpenAI has a strategic NVIDIA systems partnership. | Checks out | NVIDIA and OpenAI announced a strategic partnership to deploy at least 10 gigawatts of NVIDIA systems, with the first gigawatt planned for the second half of 2026. |
| Microsoft has an AI-infrastructure partnership or contract with NVIDIA. | Checks out | NVIDIA’s Rubin announcement identifies Microsoft Azure as an early deployer of Vera Rubin-based instances, and Microsoft has disclosed large-scale deployment of NVIDIA GPU clusters. |
| Meta has a multiyear AI-infrastructure partnership with NVIDIA. | Checks out | NVIDIA and Meta announced a multiyear, multigenerational strategic partnership spanning on-premises, cloud and AI infrastructure, including millions of Blackwell and Rubin GPUs. |
Cancel anytime
Advanced Micro Devices, Inc. (AMD) drops sharply as investors digest a strong earnings report that still failed to meet a high AI bar. Semiconductor weakness, profit-taking, and a rich valuation are pressuring the stock despite solid data center growth and an EPS beat.
Advanced Micro Devices, Inc. (AMD) rises after a strong post-earnings move, helped by an EPS beat, higher analyst targets, and renewed AI semiconductor optimism. The rally reflects investor confidence in AMD’s data center and AI growth story, though the stock’s premium valuation keeps expectations high.
Advanced Micro Devices, Inc. (AMD) drops after its fiscal Q2 2026 report as investors focus on an EPS miss, a high valuation, and expectations for bigger AI upside. Revenue and Q3 guidance were solid, but the market wanted more proof that growth will translate into stronger earnings.