The hype around AI often ignores a crucial question: does the tech actually work?
When a startup called Undetectable AI reached two million users last week (in only six months), people started taking notice. Could this be the long-awaited AI writing tool that produces…well, human writing?
The soaring startup has amassed hundreds of millions of views and impressions across TikTok, and other social media platforms like Reddit, but why?
Did they create something genuinely different? Can the tool actually write like a human and bypass AI detectors, or is it just AI hype? The AI market has shown to be well worth investing in, especially when the tech is disruptive and promising.
As you’ll see, when it comes to this brand-new startup, there is something different happening.
Undetectable AI: Promising Premise or Hype?
Undetectable AI has a core pitch that is undeniably alluring: AI-generated human-like text that slyly evades detection.
“It’s like a human wrote it,” beams Devan Leos, Chief Communications Officer. A bold claim in a booming market projected to hit a whopping $207 billion this year.
But the devil is in the details. How exactly does Undetectable AI achieve its human-like output? The company has closely guarded its technical secrets. While understandable for competitive reasons, the bold claims initially prompted my skepticism. Does its “undetectable” technology reflect true AI innovation or clever marketing hype?
Without access to the algorithmic codes and datasets, experts caution against prematurely crowning anything an “AI game-changer.” But that being said, AI-detection tools and academics alike have pushed the idea that a truly “Undetectable AI” is cause for alarm.
Since May, a few startups have popped up claiming to offer “UNDETECTABLE AI!” and promising to “BYPASS AI DETECTION!!!” But they mostly began showing up after the success of Undetectable.ai—and none have amassed the following of the titular leader in the space.
Undetectable AI Guaranteed: For Only $10 A Month
Priced between $10-$50 monthly, Undetectable AI is positioned for wide adoption from solopreneurs to big brands. Affordability spurs growth, CEO Christian Perry notes, allowing the platform to “democratize AI content creation.”
A noble goal perhaps, but again, lack of transparency raises concerns. How can the technology be offered so cheaply if its so advanced?
What about the ethical consideration of providing such a powerful tool to the masses? These are logical questions, but tools like Claude and ChatGPT offer free plans, so the small price isn’t necessarily indicative of small capabilities.
Engagement And Ethics
With 2 million monthly active users, Undetectable AI clearly provides value to many. CTO Bars Juhasz credits this viral success to the platform’s uniqueness. “It transforms robotic text into something eerily human,” he says. But is mere engagement enough? Shouldn’t advanced AI also demonstrate social responsibility?
Undetectable AI has taken a stance on ethics and content oversight. So how does it address risks like bias, misinformation, and “deepfakes” proliferating through its human-like text?
Well, for starters the platform says they prohibit cheating and misconduct–which okay, that’s a start.
However, they also focus on diversity, inclusion, and providing AI tech to economically disadvantaged groups, one of which is small businesses. In fact, Bars Juhasz the creator of Undetectable was nominated for a VentureBeat diversity award for the efforts.
Undetectable.ai also features a free AI detector on their site; no payment is ever required to use the detector in any capacity.
I can imagine this tool being used for good by allowing people who struggle to write well or communicate to use AI without sounding like they’re using AI.
Of course, the fear of misinformation spreading could also be a concern. But for spreading helpful information, a tool that can help create more trustworthy AI content could also be beneficial.
Undetectable.ai: Is It Really Worth The Hype?
Rather than coronate Undetectable AI as the next AI marvel, its meteoric rise might warrant balanced scrutiny. Its claims suggest potential, but does the product lack scientific rigor?
Its affordability may seem dubious given the advanced technology. But its public messaging emphasizes ethics, and its developer Bars Juhasz is a credentialed machine learning expert and Ph.D.
Unlike other startups, Undetectable.ai isn’t promising to do your homework. Their target audience seems to be varied. However, they offer something that no other AI-writing tool has before and at a fairly low cost.
Does this mean Undetectable AI is assuredly pure hype? No. But responsible AI requires evidence-based assessment, not blind boosterism.
Undetectable AI makes an alluring sell in a lucrative market. The claims of human-like content creation done for you (by AI) need facts to back it up.
But recent research reveals a double-edged dagger that backs the startups’ claim while also raising questions about AI-detectability.
Research: Undetectable.ai Deemed Mostly Human By Detectors
While the vast majority of tools out on the market are merely ChatGPT re-skins, and not really worth investing your time (or money) in, recent research shows that Undetectable.ai works a lot more effectively than one might suspect.
A recent study by Taloni et al. (published in Nature Eye) provides intriguing early evidence that Undetectable AI is an example of AI tools maturing in their ability to produce convincing human-like text.
- The researchers tested Undetectable AI by having it “humanize” 20 abstracts from published medical articles.
- They found the Undetectable AI versions showed low plagiarism scores around 10%, indicating unique writing.
- When the researchers used Undetectable AI’s “humanizing” feature, the AI detection score dropped significantly from 91% to just 28%.
- The readability of the abstracts also improved based on standard metrics like the Flesch Reading Ease score.
The authors note, “These findings highlight the capability of Undetectable AI to produce plagiarism-free scientific essays, and, when refined using humanizing tools, to evade AI detection.“
The main kicker in all this? The research used what some believe to be the most accurate AI detector on the market, called originality.ai. Originality says their detection tool is 95% accurate, but while it could nab regular ChatGPT texts, it was only 25% accurate when trying to detect Undetectable AI.
Verdict: Undetectable AI Might Be Worth The Hype
Undetectable AI sounds a bit nefarious initially, and “Bypassing AI detection” seems a little devious at first glance.
But the founders of this startup have demonstrated their expertise and ability to stand out—all the while pushing for ethical use cases and trying to champion disadvantaged people groups. This combined with exorbitant growth seems to signal more than just demand, and perhaps the markers of truly innovative product.
Considerable debate surrounds the ethics of advanced AI writing tools in academia and beyond. And people are on both sides of the fence. Some embrace AI, while others see it as this decade’s y2k.
Nonetheless, the recent study provides initial evidence that the latest version of Undetectable AI can mimic markers of human writing style and content fairly well when paraphrasing or rewriting text, which is how it effectively bypasses AI detectors.
While more remains to be seen, Undetectable AI appears to be making major strides in generating topical texts that are nearly indistinguishable from human-written words.
When I tried undetectable.ai, I used a few high-level tools to try and detect it. Only one of the ChatGPT outputs got flagged, and only as 29% AI. Maybe the detectors need to up their ante. Grab your popcorn because the AI vs. AI battle is just beginning to unfold.
The ultimate impact remains to be seen, but the latest evidence lends credence to Undetectable.ai not just being another over-hyped AI product.
Regardless of which side you’re on, this proactive new AI tool’s hype might be worth exploring.