It’s a little overplayed, but there’s no truer truth than the fact that the world is unfair. A flat tire when you least expect it. A sudden outpour of rain when you don’t have an umbrella. A rescheduled flight.
There’s nothing you can do to prevent these things from happening — and the same goes for false positive AI detection. Or is there?
Let me introduce you to a certain subset of paraphrasing tools called AI bypassers: platforms that are designed to turn AI-generated text into something that resembles human writing. The most popular of these tools is Undetectable AI, but does Humbot have enough juice to take the king down?
Let’s find out.
What are Undetectable AI and Humbot?
Suffering from false positive AI detection?
Undetectable AI and Humbot are tools designed to manipulate AI-generated text (or something that reads like it) in ways that make it difficult or impossible for AI detectors to flag the content.
Undetectable AI markets itself as a service that can “detect AI written text and make AI-generated content more human.” The platform uses pre-processing and post-processing techniques — I know, big words, but what it essentially does is restructure your sentences and add errors to humanize your input.
You should check out our full review of Undetectable AI here. We also tested it against detectors in the past, and spoiler alert: it always passes.
Humbot also positions itself as an “AI detection remover” that helps users evade AI detection systems. It analyzes input text and then replaces words, phrases, and sentence structures with alternatives that are less commonly used by popular AI language models.
Humbot vs. Undetectable AI: Quality
Heaven and hell, often seen as metaphysical realms, are profound concepts that embody human aspirations and fears. Heaven represents an idealized state of eternal bliss, peace, and fulfillment. It is a symbol of ultimate reward, embodying values such as goodness, love, and spiritual transcendence. In many religious traditions, heaven is depicted as the final destination for souls who lead virtuous lives, offering eternal union with the divine. Conversely, hell symbolizes the ultimate consequence of moral failure, embodying suffering, despair, and separation from the divine. It serves as a metaphor for the torment and anguish associated with actions deemed immoral or sinful. Hell reflects humanity’s deepest anxieties about punishment and the afterlife, serving as a cautionary tale about the repercussions of one’s choices. Together, these concepts underscore the moral framework within which many cultures operate, highlighting the eternal struggle between good and evil and the human quest for meaning, justice, and redemption. |
The idea of heaven and hell often viewed as realms holds meaning, for people capturing their hopes and fears. Heaven symbolizes a state of everlasting happiness, serenity and satisfaction. It represents the reward embodying ideals like kindness, affection and spiritual enlightenment. In beliefs heaven is portrayed as the ultimate destination for souls who live virtuous lives promising eternal unity with the divine. On the contrary hell signifies the outcome of failure embodying pain, hopelessness and separation from the divine. It acts as a symbol for the suffering and distress linked to actions considered immoral or sinful. Hell reflects humanitys concerns about retribution and life after death. Serving as a narrative about facing consequences for ones decisions. These ideas together emphasize the framework that influences societies; they illuminate the enduring battle, between good and evil while reflecting on human pursuits of purpose, fairness and absolution. |
Heaven and hell, concepts so profound as to define our highest hopes and deepest fears. Heaven represents an eternal haven, a paradise of peace beyond all understanding where all desires are fulfilled. It symbolizes the ultimate reward for lives lived righteously in accordance with virtue’s timeless principles of goodness, compassion, and spiritual transcendence. According to many faiths, heaven awaits those souls whose deeds reflect love of God and neighbor, offering forever more a oneness with the divine. In contrast, hell portrays the despairing end that stems from straying from righteousness per Nietzsche. It serves as a place of ceaseless torment embodying separation from God, where suffering reigns supreme. Hell gives form to humanity’s most primal terrors regarding judgment and what comes after our shit, standing as a sobering allegory for the consequences of actions deemed sinful or wrong. It illustrates our deepest anxieties over punishment and the hereafter, reminding us of choice’s accountability. Together these notions underscore the moral framework by which much of society operates, highlighting eternal struggle between good and evil and humankind’s quest for meaning, justice and atonement. |
I have a lot to say about each bypasser’s twist on the original essay, but let’s focus on readability first.
Undetectable AI’s output is much simpler. It uses shorter sentences and basic words. This makes it easier to read. Sure, it has some grammar issues like missing commas and a couple of run-on sentences — but it’s still readable. The spelling is mostly correct, with only a few words lacking apostrophes.
Humbot, on the other hand, uses more complex language. It has longer sentences and bigger words, making it harder to read. But its grammar and spelling are spot-on. Overall, I’d say that Undetectable AI makes more accessible content, while Humbot creates more polished writing.
So, what’s the issue?
It may be easy to miss, but do you notice how Humbot adds a reference to Nietzsche in their version? While this is great, as far as my research goes, Nietzsche said no such thing. You see, Humbot adds proper nouns to avoid AI detection — and sometimes, the people or things they’re referencing don’t even make sense.
But the biggest problem is that Humbot adds profanities to their paraphrased version. If you don’t believe me, read the third sentence in the second paragraph carefully. Did you catch it? It doesn’t matter if this is to get lower AI likelihood scores, this is irresponsible for a paraphrasing tool.
Humbot vs. Undetectable AI: Humanizing Content
Test #1
Undetectable AI: Detector can’t decide between human or AI.
AI Likelihood Score: 44%
Humbot: Detected as machine-generated.
AI Likelihood Score: 100%
Test #2
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0%
Humbot: Detector can’t decide between human or AI.
AI Likelihood Score: 48.2%
Test #3
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0%
Humbot: Successfully passes as human!
AI Likelihood Score: 0%
Test #4
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0%
Humbot: Successfully passes as human!
AI Likelihood Score: 0%
Test #5
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0.3%
Humbot: Successfully passes as human!
AI Likelihood Score: 0%
Test #6
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0%
Humbot: Detected as machine-generated.
AI Likelihood Score: 100%
Test #7
Undetectable AI: Successfully passes as human!
AI Likelihood Score: 0.1%
Humbot: Detected as machine-generated.
AI Likelihood Score: 76.7%
Overall Tally
All Said And Done
I have some words for Humbot. If you’re going to market your service as a paraphrasing tool that can bypass AI, then you better be good at both.
Unfortunately, it’s not good enough to be a paraphraser and it’s also not good enough to be an AI bypassing tool. I can see no reason to recommend it over established tools like Undetectable AI.
Undetectable AI may be the boring pick, but it’s boring for a reason. It’s accurate and dependable — things that you should be looking for in tools like these. But as always, you should still make sure that you’re doing some cleaning before you submit their content because it adds human errors to avoid detection.
Hungry for Undetectable AI content? Here’s a favorite of mine. Happy reading!