Picture detection has grow to be a weirdly aggressive house recently. A 12 months in the past, there have been possibly three instruments value speaking about. Now? Each different platform immediately claims it might “detect deepfakes” or “spot AI-generated media with near-perfect accuracy.”
And generally, these claims come from firms you’d by no means anticipate — like BrandWell.
BrandWell is understood for content material advertising and marketing, search engine optimisation automation, and long-form workflow instruments. They generate 7,000-word articles, automate inside linking, run full content material audits… and someplace alongside the best way, they determined to throw in an AI picture detector too.
Naturally, I needed to see the way it compares to TruthScan, a software constructed particularly for AI detection and media forensics.
Two completely completely different firms. Two very completely different objectives.
However one query: Which one really detects AI pictures accurately?
What’s TruthScan?
TruthScan is the Swiss Military suite of AI detection. The place most instruments do one factor, TruthScan has a whole lineup consisting of an AI textual content detector, AI picture detector, deepfake detector, real-time AI detector, AI voice detector, and e-mail rip-off detector.
It’s constructed for accuracy, consistency, and forensic-level detection — and the purpose is easy: catch AI-generated or manipulated content material quick and assist organizations keep forward of AI misuse.
The half we’re at this time is its AI Picture Detector. And truthfully, that is the place TruthScan shines. It’s quick, it’s assured, and it often returns a correctness rating that makes most different instruments appear like they’re guessing.
If there’s one software constructed particularly for telling actual from synthetic, it’s this one.
What’s BrandWell?
BrandWell is a content material and search engine optimisation platform — previously Content material at Scale — designed to assist entrepreneurs, businesses, and companies automate their long-form content material pipeline.
So what does that need to do with AI picture detection? Properly… technically, nothing. The AI picture detector is extra of an add-on, not one in all its core flagship instruments. It’s a full-stack content material engine.
However as a result of they already work closely with AI, and since picture era is now tied to content material workflows, they rolled out an AI Picture Detection function to assist customers confirm the authenticity of media earlier than publishing.
Query is, how nicely does it really carry out when positioned side-by-side with TruthScan?
TruthScan vs. BrandWell: Which Spots AI Higher?
Take a look at #1
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 83%
BrandWell: Detector uncertain.
AI Probability Rating: 57%
Take a look at #2
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 97%
BrandWell: Incorrectly labeled picture as human.
AI Probability Rating: 4%
Take a look at #3
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 98%
BrandWell: Incorrectly labeled picture as human.
AI Probability Rating: 6%
Take a look at #4
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 97%
BrandWell: Accurately labeled picture as AI-generated.
AI Probability Rating: 65%
Take a look at #5
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 99%
BrandWell: Incorrectly labeled picture as human.
AI Probability Rating: 24%
Take a look at #6
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 99%
BrandWell: Incorrectly labeled picture as human.
AI Probability Rating: 10%
Take a look at #7
Truthscan: Accurately labeled picture as AI-generated.
AI Probability Rating: 97%
BrandWell: Incorrectly labeled picture as human.
AI Probability Rating: 2%
Common Rating
Backside Line
After operating each instruments by means of the identical set of pictures, the hole was simply large.
TruthScan landed a 95.7% accuracy rating, which truthfully didn’t shock me. It’s constructed for this. Picture forensics, deepfake detection, media authenticity — that’s its core id. If you use a software created particularly for AI detection, you anticipate excessive efficiency, and TruthScan delivered precisely that.
BrandWell, however, scored 24% accuracy.
Is that low? Sure.
However to be honest, BrandWell was by no means designed to compete with forensic-grade detectors. Its AI picture detection software is extra of a bonus function inside a a lot bigger search engine optimisation and content-marketing engine. It’s there to assist customers sanity-check pictures earlier than publishing — to not grow to be the gold normal of authenticity verification.
So right here’s the sincere takeaway:
In the event you want a fast, built-in detector whereas engaged on content material contained in the BrandWell ecosystem, their software will get the job performed at a really primary stage. However in case you really care about whether or not a picture is AI-generated — and also you want outcomes you’ll be able to belief — TruthScan is a number of leagues forward.
