Gemini 3 Pro scores 69% trust in blinded testing up from 16% for Gemini 2.5: The case for evaluating AI on real-world trust, not academic benchmarks

VentureBeatWednesday, December 3, 2025 at 10:00:00 PM
PositiveTechnology
Gemini 3 Pro scores 69% trust in blinded testing up from 16% for Gemini 2.5: The case for evaluating AI on real-world trust, not academic benchmarks
  • Google has introduced its Gemini 3 AI model, achieving a significant increase in user trust, scoring 69% in blinded testing, compared to just 16% for its predecessor, Gemini 2.5. This evaluation, conducted by Prolific, emphasizes real-world trust over traditional academic benchmarks, highlighting the model's practical effectiveness.
  • The improved trust score for Gemini 3 is crucial for Google as it seeks to solidify its position in the competitive AI landscape, particularly against rivals like ChatGPT. This advancement not only enhances user interactions but also positions Google as a leader in ethical AI development.
  • The launch of Gemini 3 reflects a broader trend in the AI industry, where user trust and practical applications are becoming increasingly important metrics for success. As companies strive to develop AI that resonates with users, the focus on real-world performance over theoretical benchmarks may reshape how AI technologies are evaluated and adopted.
— via World Pulse Now AI Editorial System

Was this article worth reading? Share it

Recommended apps based on your readingExplore all apps
Continue Readings
Code-red in OpenAI HQ? Sam Altman Issues A Warning
NegativeTechnology
OpenAI CEO Sam Altman has declared a 'code red' for ChatGPT following a significant surge in user adoption of Google's Gemini 3, which has reportedly gained 200 million users within three months of its launch. This alarming development indicates that ChatGPT may no longer hold its position as the leading AI platform, prompting urgent internal measures at OpenAI.
Google just gave Android users several compelling reasons to stay (including this scam tool)
PositiveTechnology
Google has introduced several new features for Android 16 users, including urgent call indicators, enhanced scam protection, and pinned tabs in Chrome, aimed at improving user experience and security. These updates reflect Google's ongoing commitment to enhancing its Android platform.
India revokes order to preload smartphones with state-owned security app
NegativeTechnology
India's government has revoked its order requiring all smartphones to be pre-installed with the state-owned Sanchar Saathi cybersecurity app, following significant public backlash and privacy concerns raised by tech companies like Apple and Google. The Department of Telecommunications confirmed the reversal after the initial mandate aimed to enhance national security and combat rising cybercrime.
Nvidia’s Fat Margins Are Google and AMD’s Opportunity
NeutralTechnology
Nvidia's strong position in the AI chip market is under scrutiny as its profit margins appear vulnerable, raising concerns about its future competitiveness against rivals like Google and AMD. Recent reports indicate that Nvidia has achieved record revenues, but the pressure from competitors is intensifying.
YouTube says it will comply with Australia’s under-16s social media ban, with Lemon8 to also restrict access
NeutralTechnology
YouTube has announced its compliance with Australia's upcoming ban on social media access for individuals under 16, which is set to take effect on November 10, 2025. This decision follows a warning from Google's parent company that the laws may not effectively enhance online safety for teenagers. Communications Minister Anika Wells emphasized the platform's responsibility to ensure user safety.
Workspace Studio aims to solve the real agent problem: Getting employees to use them
PositiveTechnology
Google has made its Workspace Studio generally available, aiming to enhance employee engagement with AI agents developed by their teams. This initiative is part of a broader strategy to democratize access to AI tools within organizations, positioning Google against competitors like Microsoft and Amazon in the enterprise AI space.
Google Discover is testing AI-generated headlines and they aren't good
NegativeTechnology
Google Discover is currently testing AI-generated headlines, but early feedback indicates that the results are subpar, raising concerns about the reliability of AI in content creation. This testing phase reflects Google's ongoing efforts to integrate artificial intelligence into its services, despite the mixed outcomes observed so far.
OpenAI CEO declares “code red” as Gemini gains 200 million users in 3 months
PositiveTechnology
OpenAI CEO Sam Altman has declared a 'code red' for ChatGPT as Google’s Gemini 3 rapidly gains traction, amassing 200 million users within just three months of its launch. This shift marks a significant change in the competitive landscape of AI technologies, with Google now posing a formidable challenge to OpenAI's flagship product.