I Tested 100+ AI Tools - Here Are THE WORST

Today, I'm showing you AI products that, in our view, deliver poor value for most users. Pointless features nobody asked for, devices that don't work, and tools that are just right in the hype wave straight into your wallet. By the end of this video, you'll know exactly what to avoid and why this even exists in the first place. Let's get into it. First

up, the Humane AI pen and the Rabbit R1. If you've been following tech news, you've probably heard of these. The Humane AI pen was this little wearable device that clips onto your shirt. It cost $700 and you had to pay a monthly subscription on top of that. The idea replace your phone with a voice assistant and a tiny projector. In third-party reviews and hands-on tests, users reported heat charging quirks. The projector was often hard to see in daylight and responses felt slower than just using a phone. Before we move on, I

want to show you something that's actually been helpful while researching all this. I've been using the deep AI research inside AMS or Pro to track and analyze these AI products. Inside the platform, there is a curated AI digest that filters out the noise so I don't waste time reading every hype article. Then there's the Rabbit R1. This one was hyped as a standalone AI assistant device. It raised millions and funding, sold out in pre-orders, and then it launched. Developers demonstrated similar functionality running on Android. Reviewers argued the dedicated device added little beyond a phone. It had limited integrations at launch, inconsistent AI responses and reviews, and disappointing battery life for some users. Some reviewers even described it as a glorified toy. But here's the thing, these devices, they tried to replace something that already works perfectly fine, your phone. Your phone already has voice assistance, already has a screen, already connects to your apps. Why would you want a separate device that does less, costs more, and requires you to charge another thing every night? The answer is you wouldn't. And the market made it very clear. And then we have the AI baby stroller. Yes, this is real. A company developed an AI powered stroller with self-driving features, obstacle detection, and automatic braking. Sounds cool, right? Except for one tiny detail. The self-driving mode only works when the stroller is empty. The moment you put a baby in it, you know the entire point of a stroller, the feature shuts off for safety reasons. So, you're paying thousands of dollars for a feature you literally can't use. Oh, and speaking of unusual devices, the HushMe voice mask. This is a wearable muzzle style device that muffles your voice so you can take conference calls in public without disturbing anyone. It looks like something out of a dystopian sci-fi movie. The problem, most people aren't going around in public wearing a face mask device just to make a call. You know what else works? Finding a quiet spot or using headphones. This device solves a problem that didn't need solving and creates a much bigger one, looking absolutely ridiculous. Look, I'm always testing new AI tools to simplify

and optimize my team's workflow. For example, I spent last week testing AI website builders that all promise the same thing, build your site in minutes. Here's what actually happens. You wait 10 minutes for the AI to load. It spits out some generic template that looks like it's from 2015. And when you try to customize it, well, good luck. Half the features don't work. The other half require coding. I needed a simple portfolio site for a site project. And I was this close to just hiring someone on Fiverr. Then I found Durable, and I'm not exaggerating, built me complete professional site in 30 seconds. I typed in freelance AI consultant, hit generate, and boom, full site with multiple pages, copy that actually makes sense, built-in SEO, even a contact form. And here's the part that sold me. It's actually easy to edit. No weird drag and drop bugs, no oops, you just broke the entire layout. I can click any text, change it, move sections around, and it just works. Plus, they just launched Studio. So now you can create logo, social post, and branded visuals without jumping to Canva or paying for another subscription. Everything's in one place. Durable knows your business details and seamlessly creates the creative you need to present your business in the best way possible. Oh, and they have AI assistance currently in beta. This one helps with marketing copy. Another handles customer questions. It's like having a tiny team inside the platform. This is what AI tools should have been from the start. Fast, simple, and actually useful. Try Durable. Use my code for 30% off all plans. Link in the description. Now

let's move from consumer gadgets to the big leagues. These aren't small startups burning through seed funding. These are billiondoll corporations deploying AI systems that failed catastrophically, embarrassed their brands, and in some cases caused real problems. In May 2024, Google rolled out the early version of AI overviews, a feature that started replacing traditional search results with AI generated summaries at the top of the page. The idea was to give users instant answers without clicking through the website. Within days, that initial roll out turned into a public relations disaster. Screenshots went viral of the AI suggesting people add glue to pizza to make the cheese stick better, recommending Eden rocks for nutritional value and even telling that users might use gasoline for cooking. Why did this happen? Because in those early versions, Google's AI couldn't reliably distinguish between factual reporting and satirical joke posts on Reddit. The system scraped content from forums, treated sarcasm as legitimate advice, and confidently presented it as authoritative information. It didn't stop at silly answers, either. Some responses included genuinely dangerous misinformation. This wasn't a minor bug that slipped through testing. It exposed a fundamental weakness in that first iteration of AI overviews, a poor grasp of context, satire, and source credibility. Google had to scramble to pull back the most problematic experiences, issue statements, and rework the underlying safety systems. To be fair, since then, most of the obvious failures have been patched. Today, AI overviews usually behave far more conservatively. For many everyday queries, they're reasonably accurate. And when a request is too complex, ambiguous, or high-risisk, Google often simply doesn't show an overview at all. But that early fiasco, it's a textbook example of what happens when companies rush AI products to market, chasing hype and speed instead of treating accuracy and safety as non-negotiable. Apple entered the AI race with a feature called Apple Intelligence, which included notification summaries designed to help users stay informed by condensing news headlines and alerts. Instead, the system generated completely fabricated news. These weren't small errors or misinterpretations. They were entirely fabricated events presented as real news and attributed to credible journalistic sources like the BBC. Apple pulled the feature shortly after these incidents went public, but the damage was done. When an AI system starts inventing news and attaching the names of real media outlets to it, it doesn't just create awkward moments. It puts serious pressure on questions of ethics, trust, and responsibility. News organizations rely on their reputation to survive, and even short-lived glitches like this can blur the line between verified reporting and machine generated fiction. At the same time, this incident shows the growing pains of rolling out AI systems at scale. Amazon promoted its just walk out technology as a revolutionary AIdriven shopping experience. The pitch was simple. Walk into an Amazon Go or Amazon Fresh store, grab what you want, and leave. No checkout, no scanning, no cashiers. Computer vision and machine learning would track your items automatically and charge your account. It was framed as the future of retail, frictionless and fully automated. Later reporting revealed that behind the scenes, the system still relied heavily on human reviewers. According to multiple reports, roughly 1,000 workers in India were involved in reviewing shopping data and camera footage, double-checking a large share of transactions when the AI wasn't confident about what a customer had taken. In other words, this wasn't a clean, set it and forget it AI system, but a hybrid of automation plus a sizable human team doing validation and annotation. Critics argued that the marketing around just walk out overstated how autonomous the technology really was and blurred the line between AI powered and humanupported automation. Amazon for its part has pushed back on the idea that people were running the system in real time saying that the Indiabased teams primarily help train the machine learning models and step in to validate a minority of visits where the computer vision system isn't fully confident. Over time, Amazon has scaled back Just walk out in its larger Amazon Fresh grocery stores, shifting focus to Dash Cards, smart shopping carts that track items as you shop, while continuing to use Just Walk Out in smaller format stores, and some third-party locations. The result is a more nuanced picture than the original marketing suggested. And the whole episode has become a case study in what many now call AI washing, promoting an experience as fully autonomous AI when in reality it still depends on substantial human work behind the scenes. According to a published medical case report, a man developed brahmism, a rare and serious form of broine toxicity. He later told doctors that this decision was influenced by a conversation he'd had with Chad GBT where he was looking for ways to change his salt intake and came away with the idea of using sodium bromide instead of regular table salt. The clinicians who treated him didn't have access to his original Chad. But when they tested similar prompts themselves on an older version of the model, they found that it could mention broomemide in a matter-of-act way without strong health warnings or clear instructions to speak to a doctor first. Cases like this are exactly why medical professionals keep stressing that AI chatbots should not be treated as doctors. Open AAIS and other providers include disclaimers stating that their tools are not meant for medical diagnosis or treatment. But in practice, users may overlook those warnings. This incident is a sharp reminder that in highstakes areas like health, AI should be used at most as a supplementary tool, never as the final decision maker. Elon Musk's AI chatbot grock integrated into the Xplatform has already been at the center of several highprofile safety controversies. The system was marketed as a more unfiltered assistant with fewer content restrictions than many competitors. In practice, that looser approach led to some serious incidents. In one widely reported case, Grock generated stepbystep instructions describing how to get into a specific user's home after another user prompted it to do so. Around the same time, Grock also began producing dangerous responses, including strange and critical posts. These outputs triggered strong backlash from users, civil rights groups, and the Anti-Defamation League, who raised concerns about the chatbot's content safeguards and the risks of deploying such a system at scale. In response, X and XAI removed the offended posts, temporarily took Grock offline and updated its system prompts and filtering logic to tighten safety controls. Musk and XAI later characterized the behavior as the result of flawed configuration and an overcompliant model and said they were working on making Grock more truth seeeking and resistant to that kind of

prompts. Now let's dive into the smartarthome AI. These are products that claim to use cuttingedge AI but are really just basic sensors with a premium price tag slapped on top. There's a product on the market claiming to monitor moisture and plant health using AI. Here's what's actually happening. It operates on basic sensor feedback, not machine learning. Traditional moisture sensors have been doing this exact job for decades, and they cost about $15. This AI version, 60 to 80 bucks. You're paying a 400% markup for the word AI on the box. The pot checks if the soil is dry. That's it. No neural networks are involved. No algorithms learning your plant's behavior. It's a moisture sensor with Bluetooth connectivity and an app that sends you push notifications. Then we have ChatterBaby, an app that claims to decode baby cries into specific needs. Hunger, fatigue, discomfort. The pitch sounds great. Your baby cries, the app tells you what's wrong. Parenting gets easier, except the accuracy rates are inconsistent at best. Multiple pediatricians have publicly cautioned parents against relying on this technology. Here's why. Babies cry for overlapping complex reasons. Cry might signal hunger and tiredness simultaneously. Context matters, time of day, recent feeding schedule, diaper changes, environmental factors, an algorithm listening to audio patterns can capture that nuance. What happens when parents trust the app over their own instinct? They second-guess themselves. They ignore obvious cues because the app says something different. This isn't just unhelpful, it's potentially harmful. Firstly, the app was developed for deaf parents. So, in this case, it must be useful. But for others, better to use their parenting instincts. Pediatricians emphasize that parental observation and intuition combined with basic knowledge of infant needs are far more reliable than any app trying to interpret cry patterns. And if you thought baby translation was funny, meet me talk, the AI cat translator. This app claims to use neural networks to decode your cat's meows into understandable language. It will supposedly tell you if your cat is hungry, wants attention, or is annoyed. Here's the problem. There is no peerreed scientific validation back in any of this. The app offers generalized guesses based on audio patterns, but no credible evidence suggests AI can accurately translate cat vocalizations. Cats communicate through a combination of vocalizations, body language, ear positioning, tail movement, and context. A meow in front of an empty food ball means something different than a meow at the door. The app strips away all that context and tries to interpret sound alone. It's like trying to understand a conversation by only hearing the tone of someone's voice without knowing what they're saying or where they are. Your cat meows because it's hungry, wants to go outside, or is irritated that you're not paying attention. You don't need an app to figure that out. Finally, let's talk about robotic AI pets like Parro and Mofflin. These are plush robotic animals that mimic affection through AI powered sound and motion. They respond to touch, make cute noises, and move in ways that simulate a living pet. The pitch is emotional support without the mess, allergies, or responsibility of a real animal. The problem? Emotional substitution lacks the long-term cognitive and psychological benefits that real animals provide. Studies have consistently shown that genuine pet interaction reduces stress, lowers blood pressure, increases oxytocin levels, and improves mental health. A robotic seal making programmed sounds based on sensor input doesn't replicate any of that. There's no genuine bond, no reciprocal relationship, no unpredictability that makes real animals emotionally engaging. You're paying hundreds or even thousands of dollars for what is essentially an expensive electronic toy that mimics affection without delivering the actual psychological benefits of animal companionship. For elderly individuals in care facilities where real pets aren't feasible, these robots might offer temporary comfort, but marketing them as equivalent replacements for actual pets is misleading. Now, let's

talk about AI features that technically work but serve absolutely zero purpose. These are products that had AI slapped onto them just because AI is a buzzword that sells. Take the AI BBQ grill. It monitors temperature, adjusts heat automatically, and even suggests cooking times based on the type of meat you're grilling. And sure, that sounds convenient, but grilling is supposed to be fun. Half the point of grilling is the experience, checking the temperature, flipping the steaks or burgers, getting that perfect char. This AI grill turns it into passive weight and gain. You're not grilling anymore. You're babysitting a robot. That defeats the whole point. Will it revolutionize your workflow or finish your meal? No. will just ping you when your dinner's already burning and offer a smart recommendation. Then there's the AI toothbrush. This one costs around $300 and promises to analyze your brushing habits with AI. Functionally, its sensors and heristics tracking time and coverage features many nonAI brushes already have. My $5 toothbrush has a timer or two and it doesn't require a subscription or an app that collects data on my dental hygiene. This is peak AI washing. taken a normal product, add in some sensors, and call it AI to justify ridiculous price tag. And here's one that edges from pointless into potentially creepy. The AI companion doll, for example, Hyodol type designed for elderly companionship with connected features and cloud services. This raises privacy and dependency concerns, especially if devices handle sensitive routines or conversations. And let's not forget the L'Oreal style AI smart hairbrush. Yes, this exists. It's a hairbrush with sensors that analyze your brushing technique and hair quality, then sends the data to an app that gives you personalized hair care advice. It costs $200 for a hairbrush. You don't need AI to tell you if you're brushing your hair too hard. You can feel it. And if you really want hair care advice, you can Google it, chat GPT it for free, or

ask a stylist. This is another example of slapping AI onto a basic product and charging a premium. Now, the AI products that might work but cost so much they make no sense for most people. Exhibit A, AI bird watch and binoculars. These cost $4,800. Yes, you heard that right, $4,800 bucks. They use AI to identify bird species in real time. Display information on a built-in screen and log your sightings automatically. And sure, if you're a professional ornithologist, maybe that's useful. But here's the thing. You can get a pair of decent binoculars for $50 and use a free bird identification app on your phone. Similar outcome, fraction of the cost. This exists because AI can justify a markup, not because most people need it. Then we have AI smart mirrors that track your outfit choices and rate your outfit confidence. I'm not making this up. This mirror uses computer vision to analyze what you're wearing, compare it to fashion trends, and give you a confidence score. Who wakes up in the morning and thinks, "I really need a mirror to tell me if my jeans look good. " This is a solution in search of a problem and the only people buying it are influencers who need content. The pattern here is clear. Premium pricing does not equal premium value. AI in the name doesn't guarantee usefulness. It often means the company is justifying a markup with a buzzword and a subscription. All right, enough negativity. Let's talk about AI tools that actually make sense. Some of them

solve real problems, and genuinely improve your workflow. Take Chat GPT and other large language models. These tools are legitimately useful for drafting emails, brainstorming ideas, summarizing documents, and automating repetitive writing tasks. They save time, they're accessible, and they're getting better every month. That's real value. Look, I

spend hours every week testing these AI tools so you don't have to. And honestly, trying to keep up with the AI space while avoiding is exhausting. New tools launch every day, each one promising to revolutionize your workflow. And most of them are garbage. And that's exactly why I built my workflow around AMS or Pro. It's my home base and your home base for everything AI. And here's what that actually means in practice. First, there's the curated weekly AI digest. Every week, I get a breakdown what actually matters in AI. No fluff, no hype, just the tools, updates, and trends that are worth my time and your time. I'm not scrolling through Xfred or Reddit trying to figure out what's real. It's all filtered for me and you. Second, the platform has bite-siz lessons, over 100 of them, covering everything from AI fundamentals to advanced workflows. When you need to learn how to automate something or understand how a new model works, can jump in, watch a 5-minute lesson, and actually apply it. It's not a course you'll never finish. It's practical modular learning that fits into your schedule. Third, and this is huge, AI Master Pro has built-in tools. I'm talking about Ask AI Master, which is like having a personal AI coach, an AI art studio for visuals, prompt creator for building better prompts, and deep AI research for when I need to dig into something specific. Everything in one place. I'm not juggling five different subscriptions or remembering which tool does what. It's all there. You're not drowning in tools. You're not chasing hype. You have one hub where everything you need to learn, build, and stay current with AI is in one place. And the first 1,000 people who sign up get 24% off the annual plan. Links in the

description. Grab it before it fills up. So, why does this keep happening? Why do companies keep making pointless AI products? The answer is simple. Specifically, venture capital money. Right now, AI is the hottest thing in tech. Investors are throwing money at anything with AI powered in the pitch deck, and founders know this. So what do they do? They take a mediocre product idea, slap some AI features on it, and suddenly they're a cuttingedge AI startup worthy of millions in funding. Whether the product is useful as secondary, the story is primary. And here's the kicker. Many of the loudest founders were pushing NFTts or web 3 years ago. When that bubble popped, they pivoted to AI. Same playbook, new buzzword. They're optimizing for headlines and funding announcements, not for user value. The result, a flood of AI clutters the market and buries genuinely useful tools. So here's the takeaway. Next time you see AI powered in an ad or product launch, ask yourself two questions. First, does this solve a real problem or is it just right in the hype? And second, could I do this with tools I already have? If the answer to the first question is no, or second question is yes, then save your money. AI is incredibly powerful when it's used correctly. But slapping AI onto a product doesn't automatically make it better. In most cases, it just makes it more expensive, more complicated, more annoying. Thanks for watching. I'll see you in the next one.