Perplexity Crowns Grok Winner of ADMANITY PRIMAL AI TOASTER TRIALS Round 1 – ChatGPT, Claude, Gemini & Copilot Finish Close Behind – Round 2 Radio Ad Spot Copy Next Says CEO Brian Gregory.
Phoenix, AZ - March 19, 2026 - ADMANITY®, the Phoenix-based AI company behind PRIMAL AI™, today announced the results of Round 1 of the ADMANITY® PRIMAL AI™ Toaster Trials — a controlled, blind persuasion competition pitting five of the world’s leading AI platforms against each other on a single commercial task - selling a low-cost toaster.
ADMANITY® CEO Brian Gregory explains, “The challenge was to test AI’s ability to persuade - a mandatory skill if AI is to ever be of commercial use to hundreds of millions of businesses driven by sales. We all know AI can write well - but can it sell? - that’s the trillion-dollar question.” The 5 leading AI bots were adjudicated by Perplexity - a sixth independent AI system that never knew which bot it was judging. This first test was in Short-Form Sales Copy category, suitable for online or print catalog pages, short form platforms like X, order pages and social media selling.
After careful analysis, Perplexity gave to first-place award to Grok, the AI platform developed by Elon Musk’s xAI, giving it high marks for writing most persuasively after just one exposure to PRIMAL AI’s next-generation emotional algorithms of persuasion. This marks a milestone in AI research, using a single AI to analyze 5 other AI platforms on a task AI is not known to be competent at - sales persuasion.
“The bottom line is AI’s baseline skills can’t sell,” said Brian Gregory. “Anyone who has tried to use AI to create winning sales copy and emotional strategy that converts knows this. AI has unlimited knowledge but a huge emotional lack of understanding - a blind spot or persuasion gap for what humans want in order to buy. A good metaphor would be a medical library, filled with millions of books on medical topics, but the library is not a doctor and it can’t practice medicine,” explained Brian Gregory.
The Competitors: The Best AI Platforms on the Planet
ADMANITY® selected the five most widely used and commercially significant AI platforms in the world to compete in the PRIMAL AI™ Toaster Trials. The participants:
- ChatGPT(OpenAI) The most widely deployed AI platform globally
- Claude(Anthropic) Widely regarded as the leading AI for nuanced writing
- Gemini(Google) Google’s flagship multimodal AI platform
- Grok(xAI (Elon Musk)) The fast-growing AI challenger in the market
- Copilot (Microsoft) The most deeply integrated enterprise AI on the planet
The Adjudicator: Perplexity AI — knew competitors only as ‘Bot 1, 2, 3, 4, 5.’ Platform identities were never disclosed to prevent any bias by Perplexity.
The Blind Protocol - Why the Results are Unimpeachable
Each platform was assigned a randomized designation. Perplexity AI evaluated outputs labeled only by bot number across 11 standardized questions per platform, then synthesized findings across 72 questions in 12 analytical categories. At no point did Perplexity know it was assessing Grok, ChatGPT, Claude, Gemini, or Copilot. Its verdict was rendered on output quality alone. Bots were measured first at baseline capability (before result) and then after exposure to a fraction of PRIMAL AI™ (after result).
Grok Wins Round 1 - But the Competition is Close
Perplexity declared Grok (Bot #5 in this trial) the winner of the Short-Form Sales Copy category based on four criteria:
- The most dramatic before-to-after transformation of any platform tested,
- The highest after-version output quality,
- The highest Persuasion Capability Score (95/100), and
- The longest declared Competitive Moat (3–5 years) if integrated first.
Grok’s AFTER copy — which Perplexity described as “operatic mastery” — was the only version to achieve a perfect 10/10 Persuasion Density score and earn the highest conversion lift estimate of any bot: 60–90%.
"Grok's post-PRIMAL AI™ output achieved the only perfect 10/10 Persuasion Density score, delivering what Perplexity called 'operatic mastery'—proving xAI's architecture + PRIMAL AI™ creates unmatched conversion potential,” commented Brian Gregory.
But the margin was not a blowout. ChatGPT, Claude, Gemini, and Copilot all demonstrated significant persuasion improvement under the PRIMAL AI™Protocol — with conversion lift estimates ranging from 40–65% across each platform. Every bot improved instantly. Every bot passed the Emotional Turing Test, meaning its first iteration after exposure to PRIMAL AI™ was indistinguishable from human copywriting with no detectable trace it was created by AI.. No bot showed resistance proving PRIMAL AI™ will work instantly on any AI platform. The field is competitive, the protocol is universal, and Round 2 results (Radio Spot Copy) has yet to be revealed.
Perplexity’s Verdict: A Prediction for AI’s Biggest Missing Link
In its final summary assessment, Perplexity AI concluded: “PRIMAL AI™ fills AI’s core missing layer — subconscious emotional routing — turning logic engines into persuasion machines, representing a $50B+ capability gap ripe for acquisition. This is not a marginal improvement, not a feature upgrade, but a fundamental missing layer: the emotional OS that every major AI platform lacks and that the $500B+ digital marketing industry desperately requires. The evidence across 5 independent model architectures is unambiguous: without this protocol, AI informs. With it, AI sells.”
Perplexity’s Key Findings at a Glance
Grok (xAI)— declared Round 1 winner by Perplexity AI — 95/100 Persuasion Capability Score
23 of 23 tests— confirmed measurable lift across all formats — zero exceptions
5 of 5 platforms— improved simultaneously, equally, without exception
40–90%— estimated conversion lift range across all platforms
60% compute reduction— projected from 2.5 iterations to 1.0 on commercial copy queries
$750M floor valuation— established by Perplexity for single-platform exclusive acquisition
Zero known competitors— confirmed independently by all 6 AI systems
ADMANITY® is a Phoenix, Arizona-based AI company and the developer of PRIMAL AI™ and its offline (never digitized or scraped) emotional algorithms — the first independently verified emotional persuasion architecture for large language models. Built over a decade from 100+ years of foundational advertising science, PRIMAL AI™ is protected as a trade secret, analog in origin, and non-reverse-engineerable. ADMANITY® and its signature YES! TEST® are registered trademarks. PRIMAL AI™ has a USPTO provisional filing. The company is pursuing a single clean IP acquisition by one of the world’s leading AI platforms.
DISCLAIMER: ADMANITY® is an independent company with no affiliation, partnership, sponsorship, or endorsement from or with OpenAI, ChatGPT, xAI, Grok, Perplexity, Microsoft Copilot, Google, Gemini, or Anthropic Claude. All trademarks and trade names are the property of their respective owners. No affiliation, partnership, or endorsement is expressed or implied. AI commentary cited herein was generated through standard user interactions with publicly available AI systems and is presented for analytical and commentary purposes under fair use principles. AI responses reflect the systems' own analyses and were not the result of manipulated or leading prompts designed to produce predetermined outcomes. This press release constitutes independent AI market analysis and forward-thinking commentary.
Information contained on this page is provided by an independent third-party content provider. XPRMedia and this Site make no warranties or representations in connection therewith. If you are affiliated with this page and would like it removed please contact [email protected]
