“Google Is The Big DOG” - Gemini 3 CRUSHES ChatGPT As AI WAR Explodes

Key Concepts

Gemini 3 vs. ChatGPT: The primary focus is on the perceived leap in performance of Google's Gemini 3 over OpenAI's ChatGPT, as highlighted by Salesforce CEO Marc Benioff.
AI Ecosystem Shifts: Discussion on how the AI landscape is rapidly evolving, with a potential shift in dominance from OpenAI to Google.
Agent Technology: The importance of AI agents in powering software and performing advanced tasks is emphasized.
LLM Comparisons: Analysis of different Large Language Models (LLMs) like Gemini, ChatGPT, Perplexity, and others, focusing on their strengths and weaknesses.
Market Share and Competition: Examination of the competitive dynamics between AI companies and the potential impact on market share.
Future of Search: Speculation on how AI models will transform online search.
Investment in Self: A concluding segment promoting a business planning workshop as an investment in future success.

Gemini 3's Perceived Leap Over ChatGPT

The discussion centers on Salesforce CEO Marc Benioff's strong endorsement of Google's Gemini 3, stating it has "just blew past Chad GBT" and that he is "not going back." Benioff's tweet highlights Gemini 3's superiority in "reasoning, speed, images, video," describing it as "sharper and faster" and feeling like "the world just changed again."

However, the transcript notes Benioff's history of joining popular trends late, citing his past enthusiasm for "agents" (after they became a buzzword), the metaverse, and NFTs, which did not age well. This suggests his praise for Gemini 3 might not be an immediate indicator of a definitive market shift.

Industry Reactions and Comparisons

Sam Altman (OpenAI CEO): Congratulated Google on the Gemini 3 launch, calling it a "great model."
Andre Karpathy (Former Tesla AI Director): Had "positive early impression" of Gemini 3, calling it a "very solid daily driver potential and clearly a tier one LLM."
Patrick Collison (Stripe CEO): Noted Gemini 3's ability to build an interactive webpage summarizing genetic breakthroughs.

The transcript features a practical demonstration comparing Gemini, ChatGPT, and Perplexity. The user, Tom, uses all three for comparative analysis, particularly for understanding "agents" for business. He observes Gemini's superior "context" and "quicker" responses.

Comparative Analysis Methodology

Tom's methodology involves:

Side-by-side testing: Running the same prompts on different LLMs simultaneously in separate browsers or desktop apps.
Specific Prompt Examples:
- Asking about "consensus odds or feelings about the next head of the Fed and whether there's going to be a rate cut in December."
- Using Perplexity for "article indexes" to read original sources rather than just synthesized summaries.
- Using Julius for numerical analysis and Python output.

Live Demonstration of LLM Performance

During a live demonstration, the following observations were made:

Gemini: Demonstrated remarkable speed in generating responses.
ChatGPT: Provided a more detailed breakdown, including odds of a rate cut (75-85%) and sources, with explanations for uncertainty.
Perplexity: Quickly provided a list of 10 sources with subtitles, allowing for rapid assessment of data quality based on the websites listed.

The consensus from the demonstration was that while Gemini was significantly faster, the core information provided by all three was not drastically different for this specific query. However, the speed advantage of Gemini was noted as a significant factor.

Key Arguments and Perspectives

Competition Drives Innovation: The intense competition, particularly from Google's push with Gemini, is forcing ChatGPT to improve. This is likened to the processor wars between Intel and AMD, and the evolution of TV technology (HD to 4K).
The Future of Search: There's a strong belief that AI models like ChatGPT and Gemini will be the future of online search, potentially saving Google from losing online ad revenue.
LLM Inconsistencies: One perspective highlights that LLMs, even for sentiment tracking in finance, still exhibit inconsistencies and require back-checking.
Consolidation in the Market: It's predicted that only one or two major winners will emerge in the LLM space.
Beyond Analytical Search: A crucial point is made that the real battle is not just about the fastest or best LLM for analytical search, but about which AI powers the "agents" within enterprise software for advanced tasks. This is compared to the importance of the underlying chip powering a device.
Apple's AI Partnership Gap: Apple is criticized for not having a strong, permanent high-end AI partner to power its future AI capabilities, with Siri being noted as less advanced than current LLMs.
Market Share Dynamics: While OpenAI has the early adopter advantage, Gemini is shown to be gaining significant traction, with other players like DeepSeek and Grok also in the mix.
The "Feeling of the Mob": A prediction market (Kowi) is discussed, where people wager on which AI will be the leader. This "feeling of the mob" has historically been more accurate than traditional polling, suggesting Gemini is currently perceived as a strong contender.
Long-Term AI Evolution: It's emphasized that the AI race is still very early, and it will likely take about five years to determine the ultimate "King Kong" of the AI space.

Technical Terms and Concepts

LLM (Large Language Model): A type of artificial intelligence algorithm that uses deep learning techniques and massive datasets to understand, generate, and manipulate human language.
Agents (AI Agents): Software programs that can perform tasks autonomously or semi-autonomously on behalf of a user or another program, often leveraging LLMs for their intelligence.
SAS Multiple Bubble: Refers to a period where Software-as-a-Service (SaaS) companies were valued at extremely high multiples, often detached from their actual financial performance.
NFT Cloud: A Salesforce product offering related to Non-Fungible Tokens, which did not perform well.
TCU (Tensor Processing Unit): Google's custom-designed hardware accelerator for machine learning, used in their AI infrastructure. This is contrasted with NVIDIA's GPUs.
GPU (Graphics Processing Unit): Originally designed for graphics rendering, GPUs are highly parallel processors that have become essential for training and running AI models.
Fred Numbers: Refers to data from the Federal Reserve Economic Data (FRED) database, commonly used for economic analysis.

Data, Research Findings, and Statistics

Gemini 3 Performance: Described as "sharper and faster" in reasoning, speed, images, and video.
Rate Cut Odds (December 2025): ChatGPT estimated 75-85%; Gemini estimated 75-80%.
Market Share Trend (Hypothetical): A chart shows OpenAI (blue) declining, Gemini (red) increasing, DeepSeek (yellow) flat, Grok (green) present, and Perplexity (light blue) present.
Kowi Prediction Market (End of 2025):
- Gemini: 83%
- Grok: 9%
- Claude: 5%
- ChatGPT: Not on the list (implying a very low percentage or not considered a top contender in this specific wager).
- This represents $9.7 million wagered, with $1 bets.

Logical Connections Between Sections

The discussion flows from a specific event (Benioff's tweet) to a broader analysis of the AI market. The initial focus on Gemini 3's perceived superiority leads to comparisons with existing players like ChatGPT. This comparison then expands to include other LLMs and their specific use cases. The conversation shifts to the underlying technology and market dynamics, exploring who is powering AI agents and the potential for market consolidation. Finally, it concludes with a forward-looking perspective on the long-term evolution of AI and a call to action for personal investment in future planning.

Notable Quotes and Significant Statements

Marc Benioff (Salesforce CEO): "Gemini 3 direct competitor Chad GBT just just blew past Chad GBT. I am not going back."
Marc Benioff (on X): "Holy Penny Huff on X. I've used Chad GBT every day for three years. Just spent two hours on Gemini. I'm not going back. The leap is insane. Reasoning, speed, images, video, everything is sharper and faster. It feels like the world just changed again."
Sam Altman (OpenAI CEO): "Looks like a great model." (referring to Gemini 3)
Andre Karpathy (Former Tesla AI Director): "very solid daily driver potential and clearly a tier one LLM." (referring to Gemini 3)
Tom (User): "Wow, that's better contacts. Wow, that was a little bit wow that was quicker." (observing Gemini's performance)
Tom (User): "Competition is causing Google to say we are the big dog and we are driving hard. And Chat GBT's lead caused Google to say, hey, if you want to win the championship, you got to beat the champion."
Eric Schmidt (Former Google CEO): "You have no idea what's coming."

Synthesis/Conclusion

The core takeaway is that while OpenAI's ChatGPT has held a dominant position, Google's Gemini 3 is emerging as a significant and potentially superior competitor, evidenced by strong endorsements from industry leaders and impressive performance metrics in speed and capabilities. The AI landscape is highly dynamic, with rapid advancements and intense competition driving innovation. The future of AI will likely be shaped by the underlying technology powering advanced agents and tools, rather than solely by the speed of analytical search. While Gemini's rise is notable, the long-term dominance in the AI space is still uncertain, with significant evolution expected over the next five years. The discussion also highlights the importance of strategic investment in oneself and one's plans for future success.