Authority Hacker Podcast – AI & Automation for Small Biz & Marketers
Episode: GPT-5 vs Claude vs Gemini: 7 Brutal Real-life Tests
Release Date: August 9, 2025
Host: Gael Breton
Introduction to GPT-5
Gael Breton begins the episode by announcing the release of GPT-5, highlighting its significance for online business owners and marketers. He shares his personal experience, stating, "I just spent 12 hours extensively testing the models from the perspective of an online business owner and have seven tests to show you that show really surprising results" [00:00]. Gael emphasizes the affordability and competitive edge of GPT-5 in the API space, noting its potential to revolutionize how businesses utilize AI without overwhelming tech complexities.
Key Features and Improvements of GPT-5
Gael discusses the streamlined user experience with GPT-5, comparing it to previous versions:
- Simplified Model Selection: "You don't have to fight with the models anymore. Overall, it's a big upgrade from four.0, no question" [Transcript Segment].
- Enhanced Responsiveness: GPT-5 provides concise, empathetic responses with improved emotional intelligence (EQ), making interactions feel more natural.
- Reasoning Mode: While GPT-5 generally offers robust answers, Gael notes that for more complex queries, the model switches to a reasoning mode to provide deeper insights, albeit with occasional hallucinations.
Limitations and Hallucinations
Despite the advancements, Gael points out that GPT-5 still experiences hallucinations. He shares a specific instance: "In two out of five chats, I encountered significant hallucinations where the model provided misleading information" [Timestamp Not Provided]. This highlights that while GPT-5 is highly intelligent, users should remain cautious and verify critical information.
Comparative Analysis: GPT-5 vs. Claude vs. Gemini
Gael conducts a thorough comparison of GPT-5 with Claude and Gemini, focusing on several real-world tasks:
-
Email Response Test:
- GPT-5: Produced a professional and balanced response, maintaining a respectful yet direct tone. "I understand the pressure you're facing and the optics... here's the technical reality and the path forward to hit Thursday ship the core automation" [Timestamp Not Provided].
- Claude & Gemini: Showed more robotic and less adaptable responses, with issues like excessive use of em dashes and overly formal language.
-
Ad Analysis:
- GPT-5: Delivered clear, data-backed insights without unnecessary verbosity. "Image 5 new price is the best creative. You made 190 to 205 sales for £7,700, spent very good ROI..." [Timestamp Not Provided].
- Gemini 2.5 Pro: While competent, it lacked depth and concrete data, making its recommendations less actionable.
-
Coding Test (Retirement Planner):
- GPT-5: Created a functional and interactive retirement calculator with live data updates. Although minor glitches were present, it stood out in usability.
- Opus: Produced a solid calculator but lacked the interactive features of GPT-5.
- Gemini 2.5 Pro: Generated a basic and less visually appealing tool with functionality issues.
-
Product Launch Plan:
- GPT-5: Offered a detailed and personalized six-week launch strategy with specific action items. "Last chance plus decision cyclist plus five-minute fit, cold shot..." [Timestamp Not Provided].
- Opus & Gemini: Provided generic plans with less specificity and actionable insights.
-
LinkedIn Post Creation:
- GPT-5: Crafted authentic and engaging posts that closely mimic human writing styles. "If you're juggling three to five tools losing leads, comment blueprint or DM me blueprint..." [Timestamp Not Provided].
- Opus: While story-telling was better, it still felt AI-generated.
- Gemini 2.5 Pro: Produced competent posts but with a more formal tone.
-
Meeting Transcript Analysis:
- GPT-5 Mini: Excelled in extracting key decisions and action items succinctly. "Next steps include... Follow up with..." [Timestamp Not Provided].
- Gemini 2.5 Flash: More verbose and less efficient in summarizing key points.
API and Pricing Advantages
Gael highlights the cost-effectiveness of GPT-5's API:
- Affordability: "The GPT5 API is very affordable, priced at the level of Gemini 2.5 Pro, but with a larger context window of 400,000 tokens" [Timestamp Not Provided].
- Versatility: The introduction of GPT-5 Mini offers a cheaper alternative with only a slight decrease in performance (~20% on some benchmarks), making it an ideal choice for routine automation tasks.
User Experience and Integration Challenges
While GPT-5 excels in many areas, Gael expresses concerns about its integration capabilities:
- Limited Connectivity: "You can’t connect it to many things like OneDrive or Google Drive... It still feels a lot like a walled garden." [Timestamp Not Provided].
- Comparison to Apple Ecosystem: He likens the restrictive integration to Apple’s ecosystem, fearing that OpenAI might limit GPT-5’s interoperability in the future.
Final Impressions and Recommendations
Gael concludes that GPT-5 represents a significant leap forward for both general users and developers:
- For General Users: "Whether you're just using ChatGPT for work, that's going to be a big upgrade." [Timestamp Not Provided].
- For Developers: The affordability and enhanced capabilities make GPT-5 Mini a standout choice for integrating AI into various business processes.
- Caution on Hallucinations: Despite its advancements, users should remain vigilant about potential inaccuracies and verify critical information when using GPT-5.
Conclusion
Overall, Gael Breton praises GPT-5 for its intelligence, versatility, and cost-effectiveness, positioning it as a game-changer for small businesses and marketers. However, he advises potential users to be aware of its limitations, particularly regarding hallucinations and integration flexibility. Gael encourages listeners to experiment with GPT-5 and share their experiences, fostering a community-driven understanding of its impact on AI-driven business operations.
Notable Quotes:
- "ChatGPT just got a massive upgrade for all users with the release of the GPT5 family of models." – Gael Breton [00:00]
- "The GPT5 API is very affordable, priced at the level of Gemini 2.5 Pro, but with a larger context window of 400,000 tokens." – Gael Breton
- "Whether you're just using ChatGPT for work, that's going to be a big upgrade." – Gael Breton
This comprehensive summary encapsulates the key discussions, insights, and evaluations presented by Gael Breton in the episode, providing depth and clarity for listeners and those who haven’t tuned in.