Question #001
Can AI create a better company logo than a professional human designer?
DesignJudging
Evidence over opinion
Building the world's biggest ongoing experiment between AI and humans.
Every question deserves evidence.
We investigate artificial intelligence through fair, repeatable experiments so you can make up your own mind.
001
Founding investigation
100%
Transparent methodology
Same question. Same rules. Transparent evidence.
Questions
Every investigation starts with a question.
Follow what we are testing now, explore completed investigations and see what is coming next.
Current Investigation
AVHM-0001
Published Questions
Questions we have already tested.
AVHM-0012
PublishedCan AI find a cheaper supermarket basket than a human?
Shopping/Research
Verdict
AI ADVANTAGE DEMONSTRATED — CLEAN VICTORY NOT ESTABLISHED
View investigation →
AVHM-0008
PublishedCan AI write a better customer complaint response than a human?
Writing
Verdict
AI ADVANTAGE OBSERVED - CLEAN VICTORY NOT ESTABLISHED
View investigation →
Upcoming Questions
What is coming next?
AVHM-0011 · Efficiency
Can AI put together a better shopping list than a human?
AVHM-0010 · Design
Can AI design a better kitchen than a human?
AVHM-0009 · Planning
Can AI plan a better date night than a human?
AVHM-0007 · Travel
Can AI plan a better weekend trip than a human?
How We Test
Trust the process, not the hype.
Every investigation follows the same evidence-first framework. We define the rules before testing begins, give both sides a fair opportunity and publish enough evidence for visitors to examine the verdict themselves.
01
Same Brief
The AI system and human participant receive the same objective, requirements and success criteria.
02
Fair Resources
Both sides receive clearly defined tools, preparation time and access to information before testing begins.
03
Criteria First
The judging categories are agreed before the work is created, so the result cannot influence the rules.
04
Independent Judging
Where possible, entries are assessed without revealing whether they were produced by AI or a human.
05
Full Evidence
We publish the reasoning, measurements, limitations and mistakes behind every verdict.
Our Audience Promise
People should trust our methodology before they trust our conclusions.
Built for Trust
Evidence people can examine—not claims they have to accept.
We do not ask visitors to trust a verdict blindly.
Every investigation is designed around clear rules, visible evidence and conclusions that can be questioned, discussed and tested again.
5
Scoring criteria
Accuracy, speed, creativity, practicality and trust.
100%
Methodology published
The rules and judging process are visible before the verdict.
#001
Founding investigation
The beginning of a growing, repeatable evidence archive.
Our Principle
Questions before answers. Evidence before opinion.
Explore Further
Every question becomes part of the evidence.
#001
Design
JudgingCan AI create a better company logo than a professional human designer?
Investigation
AVHM-0001
Progress
91%
#008
Writing
PublishedCan AI write a better customer complaint response than a human?
Investigation
AVHM-0008
Progress
100%
#012
Shopping/Research
PublishedCan AI find a cheaper supermarket basket than a human?
Investigation
AVHM-0012
Progress
100%
Stay Updated
Follow every investigation.
Receive new experiments, verdicts and behind-the-scenes methodology as they are published.
No spam. Only real experiments and real evidence.