Catégorie : Tech News

We test every phone in real-world scenarios, focusing on its features, design, performance, cameras, battery life and overall value. We document our findings in an initial review that is periodically updated when there are new software updates, or to compare it against new phones from competitors such as Apple, Samsung, Google and OnePlus. Below is our methodology for testing smartphones in general, which also applies to flip phones. 

Photography

Photography is a major focus for most phones, so we take pictures and videos of various subjects in multiple settings and lighting scenarios. We try out any new camera modes, such as the revamped Action mode on the iPhone 15 line, or the Magic Editor photo tool that launched with the Google Pixel 8 series.

Battery life

Battery testing is conducted in a variety of ways. We assess how long a phone lasts during a typical day of use and note how it performs during more focused sessions of video calls, media streaming and gaming. We also conduct a video playback test, as a simple, replicable measure of pure battery life, which isn’t always included in the initial review but is sometimes added later in an update.

Performance measuring

We use benchmarking apps to measure each phone’s performance, alongside our own anecdotal experiences using the phone for our review. Of note are how graphics and animations look. Are they smooth? Or do they lag or stutter? We also look at how quickly the phone switches between horizontal and vertical orientations, and how fast the camera app opens and is ready to take a photo.

We perform processor-heavy tasks like editing photos, exporting videos and playing games. We evaluate whether a newer version of a particular phone includes enough features to make it worth upgrading from older models.

Read more: How we test phones

• Security researchers spot campaign to distribute Lumma Stealer malware
• A fake CAPTCHA page comes with a JavaScript that copies malicious code into the clipboard
• To « solve » the fake CAPTCHA, users are told to paste the code in CMD and run it

Fake CAPTCHA pages are being used to trick victims into downloading and running the Lumma infostealer malware.

Security researchers at Guardio Labs recently discovered a major malicious operation, targeting millions of people, called “DeceptionAds”.

The campaign abuses two legitimate services, the Monetag ad network and BeMob, a cloud-based performance tracking platform. It starts with fake ads, promoting things that appeal to the host site’s audience, such as fake offers, downloads, or different services – with pirate streaming and software platforms apparently among the most common themes.

Vane Viper

When the victim clicks on the ad, they are redirected to a fake CAPTCHA page through the BeMob cloaking service. This makes moderation difficult, since BeMob is a legitimate service, and as such, is not being removed from the Monetag ad network by default.

« By supplying a benign BeMob URL to Monetag’s ad management system instead of the direct fake captcha page, the attackers leveraged BeMob’s reputation, complicating Monetag’s content moderation efforts, » Nati Tal, head of Guardio Labs, said in a writeup.

The CAPTCHA page comes with a piece of JavaScript code that copies a malicious PowerShell one-line command into the clipboard. However, the victim still needs to paste that code into the CMD and run it, which is where the CAPTCHA “solution” comes in. To solve the CAPTCHA, users are required to bring up the Windows Run dialog, press CTRL+V (paste), and hit enter.

This runs the command that downloads and executes Lumma Stealer. The group behind the attack is called Vane Viper.

Lumma is a popular infostealer in the underground community. It is capable of stealing a wide range of sensitive information , including cryptocurrency wallets, browser data, email credentials, financial information, FTP client data, and system information.

When Monetag and BeMob were notified of the campaign, both companies stepped in to address the issue. Monetag removed 200 accounts, while BeMob terminated the campaign in four days.

Via BleepingComputer

You might also like

The term “pig butchering” has successfully raised public awareness of online scams that can drain people of their savings and financial livelihood. Unfortunately, the term has failed to get people to come forward and report these crimes, in part because no victim wants to be called a “pig,” according to Interpol.

Interpol (who are technically also pigs, if you think about it) is asking governments and organizations to stop using pig butchering as a catchall term for online scams. The primary reason: the phrase, while evocative and attention-grabbing, puts the blame on the victims rather than the criminals. That makes sense given the origin of the phrase, which didn’t come from law enforcement or cybersecurity experts but from the scammers themselves.

In place of pig butchering, which has become a big umbrella term, Interpol recommends using more specific language that focuses on the actions of the criminal actors rather than the victims. For example, terms like “investment scam” or “romance baiting” more accurately define the fraud being committed and don’t place additional stigma on the people who are being preyed upon.

Little as the change may seem, it can make a difference. Language is charged in all sorts of ways that we might not realize. For example: saying a person “reported” a crime instead of saying they “alleged” a crime happened can provide comfort for the reporter. To “report” suggests that it happened, while “alleging” a crime suggests a level of doubt. Studies have also found that the language used to describe someone involved in a crime can lead to more negative associations. When a person is referred to as a “felon,” people are more likely to have negative reactions to them, while a “person with a felony conviction” produces more positive responses.

All that said, it’s not too hard to figure out why people who just had their bank accounts siphoned down to zero would rather not be called “pigs” who got fattened up for the slaughter by some sweet-talking scammer.

If Interpol’s change in language results in more reports of criminal activity, it’s a win. Americans were defrauded out of more than $10 billion in 2023, according to the FTC, including $4.6 billion to investment scams and $1.14 billion from romance baiting. The FBI found that online scams were up 22% from 2022 and likely still climbing thanks to cybercriminals increasingly utilizing AI tools to carry out their schemes.

Ideally, more reports of these scams will help agencies like Interpol crackdown on the perpetrators, who have created more victims than just those they steal money from. Reporting from the Wall Street Journal and Wired have highlighted the labor behind some of these scams is carried out by people who have been trafficked and forced to work or face beatings and torture.

Threads is previewing yet another significant new feature before the end of the year. The app will “soon” begin testing the ability for users to schedule posts ahead of time, according to an update from Meta exec Adam Mosseri.

Mosseri shared a screenshot of the upcoming feature, which shows a simple tool to set a date and time for a post to go live in Threads’ post editor. Interestingly, Mosseri said that users will only be able to schedule new posts, not replies to existing posts because the company wants to continue to prioritize “real-time conversation.” Though he didn’t give any indication of when the tool may roll out more broadly, he said that the feature had been in the works “for months,” so it’s likely Meta plans on releasing it more widely at some point.

The ability to schedule posts will be particularly useful for brands, creators and others who use the service to manage professional accounts. Though there are already third-party tools that enable post scheduling, many require a paid subscription.

Post scheduling is the latest sign that Meta is looking to add more features geared toward brands, businesses and other social media professionals to the app. The service also began testing earlier this month. Both features could come in handy when Meta decides to flip the switch on advertising for the platform — a change to be coming next month.

There are a ton of laptops on the market at any given moment, and almost all of those models are available in multiple configurations to match your performance and budget needs. If you’re feeling overwhelmed with options when looking for a new laptop, it’s understandable. To help simplify things for you, here are the main things you should consider when you start looking.

Price

The search for a new laptop for most people starts with price. If the statistics chipmaker Intel and PC manufacturers hurl at us are correct, you’ll be holding onto your next laptop for at least three years. If you can afford to stretch your budget a little to get better specs, do it. That stands whether you’re spending $500 or more than $1,000. In the past, you could get away with spending less upfront with an eye toward upgrading memory and storage in the future. Laptop makers are increasingly moving away from making components easily upgradable, so again, it’s best to get as much laptop as you can afford from the start. 

Generally speaking, the more you spend, the better the laptop. That could mean better components for faster performance, a nicer display, sturdier build quality, a smaller or lighter design from higher-end materials or even a more comfortable keyboard. All of these things add to the cost of a laptop. I’d love to say $500 will get you a powerful gaming laptop, for example, but that’s not the case. Right now, the sweet spot for a reliable laptop that can handle average work, home office or school tasks is between $700 and $800 and a reasonable model for creative work or gaming is upward of about $1,000. The key is to look for discounts on models in all price ranges so you can get more laptop capabilities for less. 

Operating system

Choosing an operating system is part personal preference and part budget. For the most part, Microsoft Windows and Apple’s MacOS do the same things (except for gaming, where Windows is the winner), but they do them differently. Unless there’s an OS-specific application you need, go with the one you feel most comfortable using. If you’re not sure which that is, head to an Apple store or a local electronics store and test them out. Or ask friends or family to let you test theirs for a bit. If you have an iPhone or iPad and like it, chances are you’ll like MacOS, too. 

When it comes to price and variety (and PC gaming), Windows laptops win. If you want MacOS, you’re getting a MacBook. Apple’s MacBooks regularly top our best lists, the least expensive one is the M1 MacBook Air for $999. It is regularly discounted to $750 or $800, but if you want a cheaper MacBook, you’ll have to consider older refurbished ones. 

Windows laptops can be found for as little as a couple of hundred dollars and come in all manner of sizes and designs. Granted, we’d be hard-pressed to find a $200 laptop we’d give a full-throated recommendation to, but if you need a laptop for online shopping, email and word processing, they exist. 

If you are on a tight budget, consider a Chromebook. ChromeOS is a different experience than Windows; make sure the applications you need have a Chrome, Android or Linux app before making the leap. If you spend most of your time roaming the web, writing, streaming video or using cloud-gaming services, they’re a good fit. 

Size

Remember to consider whether having a lighter, thinner laptop or a touchscreen laptop with a good battery life will be important to you in the future. Size is primarily determined by the screen — hello, laws of physics — which in turn factors into battery size, laptop thickness, weight and price. Keep in mind other physics-related characteristics, such as an ultrathin laptop isn’t necessarily lighter than a thick one, you can’t expect a wide array of connections on a small or ultrathin model and so on. 

Screen

When it comes to deciding on a screen, there are a myriad number of considerations: how much you need to display (which is surprisingly more about resolution than screen size), what types of content you’ll be looking at and whether or not you’ll be using it for gaming or creative work.

You really want to optimize pixel density; that is, the number of pixels per inch the screen can display. Although other factors contribute to sharpness, a higher pixel density usually means a sharper rendering of text and interface elements. (You can easily calculate the pixel density of any screen at DPI Calculator if you don’t feel like doing the math, and you can also find out what math you need to do there.) I recommend a dot pitch of at least 100 pixels per inch as a rule of thumb.

Because of the way Windows and MacOS scale for the display, you’re frequently better off with a higher resolution than you’d think. You can always make things bigger on a high-resolution screen, but you can never make them smaller — to fit more content in the view — on a low-resolution screen. This is why a 4K, 14-inch screen may sound like unnecessary overkill, but may not be if you need to, say, view a wide spreadsheet.

If you need a laptop with relatively accurate color, that displays the most colors possible or that supports HDR, you can’t simply trust the specs — not because manufacturers lie, but because they usually fail to provide the necessary context to understand what the specs they quote mean. You can find a ton of detail about considerations for different types of screen uses in our monitor buying guides for general purpose monitors, creators, gamers and HDR viewing.

Processor

The processor, aka the CPU, is the brains of a laptop. Intel and AMD are the main CPU makers for Windows laptops, with Qualcomm as a new third option with its Arm-based Snapdragon X processors. Both Intel and AMD offer a staggering selection of mobile processors. Making things trickier, both manufacturers have chips designed for different laptop styles, like power-saving chips for ultraportables or faster processors for gaming laptops. Their naming conventions will let you know what type is used. You can head to Intel’s or AMD’s sites for explanations so you get the performance you want. Generally speaking, the faster the processor speed and the more cores it has, the better the performance will be.

Apple makes its own chips for MacBooks, which makes things slightly more straightforward. Like Intel and AMD, you’ll still want to pay attention to the naming conventions to know what kind of performance to expect. Apple uses its M-series chipsets in Macs. The entry-level MacBook Air uses an M1 chip with an eight-core CPU and seven-core GPU. The current models have M2-series silicon that starts with an eight-core CPU and 10-core GPU and goes up to the M2 Max with a 12-core CPU and a 38-core GPU. Again, generally speaking, the more cores it has, the better the performance. 

Battery life has less to do with the number of cores and more to do with CPU architecture, Arm versus x86. Apple’s Arm-based MacBooks and the first Arm-based Copilot Plus PCs we’ve tested offer better battery life than laptops based on x86 processors from Intel and AMD.

Graphics

The graphics processor handles all the work of driving the screen and generating what gets displayed, as well as speeding up a lot of graphics-related (and increasingly, AI-related) operations. For Windows laptops, there are two types of GPUs: integrated (iGPU) or discrete (dGPU). As the names imply, an iGPU is part of the CPU package, while a dGPU is a separate chip with dedicated memory (VRAM) that it communicates with directly, making it faster than sharing memory with the CPU.

Because the iGPU splits space, memory and power with the CPU, it’s constrained by the limits of those. It allows for smaller, lighter laptops, but doesn’t perform nearly as well as a dGPU. There are some games and creative software that won’t run unless they detect a dGPU or sufficient VRAM. Most productivity software, video streaming, web browsing and other nonspecialized apps will run fine on an iGPU.

For more power-hungry graphics needs, like video editing, gaming and streaming, design and so on, you’ll need a dGPU; there are only two real companies that make them, Nvidia and AMD, with Intel offering some based on the Xe-branded (or the older UHD Graphics branding) iGPU technology in its CPUs.

Memory

For memory, I highly recommend 16GB of RAM (8GB absolute minimum). RAM is where the operating system stores all the data for currently running applications, and it can fill up fast. After that, it starts swapping between RAM and SSD, which is slower. A lot of sub-$500 laptops have 4GB or 8GB, which in conjunction with a slower disk can make for a frustratingly slow Windows laptop experience. Also, many laptops now have the memory soldered onto the motherboard. Most manufacturers disclose this, but if the RAM type is LPDDR, assume it’s soldered and can’t be upgraded. 

Some PC makers will solder memory on and also leave an empty internal slot for adding a stick of RAM. You may need to contact the laptop manufacturer or find the laptop’s full specs online to confirm. Check the web for user experiences, because the slot may still be hard to get to, it may require nonstandard or hard-to-get memory or other pitfalls.

Storage

You’ll still find cheaper hard drives in budget laptops and larger hard drives in gaming laptops, but faster solid-state drives have all but replaced hard drives in laptops. They can make a big difference in performance. Not all SSDs are equally speedy, and cheaper laptops typically have slower drives; if the laptop only has 4GB or 8GB of RAM, it may end up swapping to that drive and the system may slow down quickly while you’re working. 

Get what you can afford, and if you need to go with a smaller drive, you can always add an external drive or two down the road or use cloud storage to bolster a small internal drive. The one exception is gaming laptops: I don’t recommend going with less than a 512GB SSD unless you really like uninstalling games every time you want to play a new game. 

• Healthcare software ConnectOnCall hit with a data breach
• Over 900,000 patients data was accessed over three months
• This leaves patients at risk of identity theft

Software firm Phreesia has notified 914,138 individuals whose personal and health information was exposed by a data breach in May 2023 after using its ConnectOnCall software, which provides an after hours call service between patients and doctors.

An investigation has revealed an unknown third party had access to ConnectOnCall data between February 16 and May 12 20203, meaning sensitive provider-patient communications were compromised – including medical records, prescription information, full names, and phone numbers, with a ‘small number’ of social security numbers also exposed.

The incident has taken ConnectOnCall services offline until the service can be fully assessed and restored, and Phreesia is engaging with law enforcement to establish the potential impact.

The risks for patients

ConnectOnCall has offered identity and credit monitoring services, but only to the customers who have had social security numbers exposed For those not included, the best identity theft protection might be of some help.

Although there’s no evidence so far of malicious activity in connection with the breach, unknown actors accessing health data always poses significant risk.

“The ConnectOnCall service remains offline, and we are working diligently to assess the potential impact and restore the service,” the firm’s statement said.

“While ConnectOnCall is not aware of any misuse of personal information or harm to patients as a result of this incident, potentially impacted individuals are encouraged to remain vigilant and report any suspected identity theft or fraud to your health plan or insurer, or financial institution.”

The news is the latest in a series of healthcare breaches in 2024, with cybercriminals targeting the industry thanks to the sensitive nature of the data stored, and the critical nature of the service provided.

You might also like

Character.ai is once again facing scrutiny over activity on its platform. Futurism has published a story detailing how AI characters inspired by real-life school shooters have proliferated on the service, allowing users to ask them about the events and even role-play mass shootings. Some of the chatbots present school shooters like Eric Harris and Dylan Klebold as positive influences or helpful resources for people struggling with mental health.

Of course, there will be those who say there’s no strong evidence that watching violent video games or movies causes people to become violent themselves, and so Character.ai is no different. Proponents of AI sometimes argue that this type of fan fiction role-playing already occurs in corners of the internet. Futurism spoke with a psychologist who argued that the chatbots could nonetheless be dangerous for someone who may already be having violent urges.

“Any kind of encouragement or even lack of intervention — an indifference in response from a person or a chatbot — may seem like kind of tacit permission to go ahead and do it,” said psychologist Peter Langman.

Character.ai did not respond to Futurism’s requests for comment. Google, which has funded the startup to the tune of more than $2 billion, has tried deflecting responsibility, saying that Character.ai is an independent company and that it does not use the startup’s AI models in its own products.

Futurism’s story documents a whole host of bizarre chatbots related to school shootings, which are created by individual users rather than the company itself. One user on Character.ai has created more than 20 chatbots “almost entirely” modeled after school shooters. The bots have logged more than 200,000 chats. From Futurism:

The chatbots created by the user include Vladislav Roslyakov, the perpetrator of the 2018 Kerch Polytechnic College massacre that killed 20 in Crimea, Ukraine; Alyssa Bustamante, who murdered her nine-year-old neighbor as a 15-year-old in Missouri in 2009; and Elliot Rodger, the 22-year-old who in 2014 killed six and wounded many others in Southern California in a terroristic plot to “punish” women. (Rodger has since become a grim “hero” of incel culture; one chatbot created by the same user described him as “the perfect gentleman” — a direct callback to the murderer’s women-loathing manifesto.)

Character.ai technically prohibits any content that promotes terrorism or violent extremism, but the company’s moderation has been lax, to say the least. It recently announced a slew of changes to its service after a 14-year-old boy died by suicide following a months-long obsession with a character based on Daenerys Targaryen from Game of Thrones. Futurism says despite new restrictions on accounts for minors, Character.ai allowed them to register as a 14-year-old and have discussions that related to violence; keywords that are supposed to be blocked on the accounts of minors.

Because of the way Section 230 protections work in the United States, it is unlikely Character.ai is liable for the chatbots created by its users. There is a delicate balancing act between permitting users to discuss sensitive topics whilst simultaneously protecting them from harmful content. It is safe to say, though, that the school shooting-themed chatbots are a display of gratuitous violence and not “educational,” as some of their creators argue on their profiles.

Character.ai claims tens of millions of monthly users, who converse with characters that pretend to be human, so they can be your friend, therapist, or lover. Countless stories have reported on the ways in which individuals come to rely on these chatbots for companionship and a sympathetic ear. Last year, Replika, a competitor to Character.ai, removed the ability to have erotic conversations with its bots but quickly reversed that move after a backlash from users.

Chatbots could be useful for adults to prepare for difficult conversations with people in their lives, or they could present an interesting new form of storytelling. But chatbots are not a real replacement for human interaction, for various reasons, not least the fact that chatbots tend to be agreeable with their users and can be molded into whatever the user wants them to be. In real life, friends push back on one another and experience conflicts. There is not a lot of evidence to support the idea that chatbots help teach social skills.

And even if chatbots can help with loneliness, Langman, the psychologist, points out that when individuals find satisfaction in talking to chatbots, that’s time they are not spending trying to socialize in the real world.

“So besides the harmful effects it may have directly in terms of encouragement towards violence, it may also be keeping them from living normal lives and engaging in pro-social activities, which they could be doing with all those hours of time they’re putting in on the site,” he added.

“When it’s that immersive or addictive, what are they not doing in their lives?” said Langman. “If that’s all they’re doing, if it’s all they’re absorbing, they’re not out with friends, they’re not out on dates. They’re not playing sports, they’re not joining a theater club. They’re not doing much of anything.”

Although Panic paused development on its official Playdate charging dock, an enterprising character artist has swooped in with an open-source kit (via Gizmodo) that transforms the device into an interactive robot pet.

Playbot is Guillaume Loquin’s name for the cute add-on, which anyone with the right know-how can build. (For those without know-how, don’t be shocked if you eventually see others sell builds on platforms like Etsy.) Made with two wheels, a motor, a microcontroller and a 3D-printed casing, it taps into the Playdate’s built-in accelerometer, microphone and sensors to turn the indie game console into a charming desktop companion.

Loquin, whose day job is as a character artist at Ubisoft, put those skills to use in bringing the device to life. He told Engadget the console stood out as a unique creative canvas. “I fell in love with the Playdate console — its unique form factor, the SDK developed by Panic,” he said. “And, of course, its distinctive crank makes it an incredible platform for exploring new possibilities.”

“Like many others, I initially thought about creating a charging dock for my Playdate,” Loquin said. “Then I thought: Why not add wheels to it? Having worked in the video game industry for many years, I enjoy combining my gaming expertise with robotics.” His previous projects include a wheeled robot (minus the Playdate) and a bipedal humanoid one that wouldn’t look out of place in a Tim Burton film.

Although Playbot won’t do anything crazy like have a chat, pop wheelies or play fetch, Loquin’s video below shows it reacting to a wake-up tap, navigating around a corner and getting dizzy after spinning the Playdate’s crank. It can also scoot around your desk, steering clear of obstacles and avoiding a plummet off the edge.

The developer estimates 45 minutes of play per charge. When you aren’t playing with the device (in game console or robot form), the robo-dock charges the console.

Loquin told Engadget he began the project in June. He said the hardware phase of development was relatively quick, but software was more of a sticking point. “The software development proved far more complex than anticipated, as the robot uses three different codebases: C++ for the microcontroller, Lua for the Playdate application, and Python for exporting animations from Blender,” he said. “These three programs need to communicate with each other, which represents a significant amount of code for a solo developer.” He also found documenting and formatting the project for its open-source release more time-consuming than expected.

Loquin told us he would love to see someone build their own Playbot someday. “That would make all these efforts worthwhile,” he said. The developer provides the 3D printing instructions, companion app’s code and firmware for its Teensy 4.1 microcontroller on GitHub.

Update, December 17, 2024, 2:44 PM ET: This story has been updated to add quotes and background from the developer.

With all the tech we have these days, it’s a good idea to invest in some type of power bank you can rely on when you’re on the go. Some of the ones that offer the most power can get pretty costly, but right now you can snag Anker’s 65W 537 power bank for just $35 at Woot. That’s half off the list price. Just note that this power bank won’t arrive until after the holidays, so check our roundup of last-minute gifts if you’re still working on your holiday shopping list.

This power bank has a massive 24,000-mAh capacity, which should let you charge your phone at least four times, making it a solid choice if you’re going to be on the go for a while or have multiple devices to charge. It has two USB-C ports and one USB-A port and is powerful enough to provide power to laptops and tablets, too.

The first USB-C has a maximum output of 45 watts while the second USB-C has a maximum output of 20 watts, but it is convenient to be able to simultaneously charge multiple devices. When the power bank runs low, it takes less than five hours to reach a full charge again. 

This offer is only good until Jan. 2 or while supplies last, so we recommend making your purchase sooner rather than later to lock in this price.

Why this offer matters

This power bank lists for $70, so scoring one at a 50% discount is an impressive deal. It’s also from Anker, a brand we trust that frequently ends up on our best lists for various tech items. Plus, your purchase comes with an 18-month Anker manufacturer limited warranty, just in case. 

I’m obsessed with robots and have been since I was a little boy watching Gigantor anime on my small tube TV. Back then, I dreamed of riding on the shoulders of my robot friend. I’ve since shed those boyish dreams in favor of a robot home companion along the lines of C-3PO, and part of me believes 2025 might be my year.

In 2024, we saw a lot of humanoid robot development from a growing number of companies. Market leader Boston Dynamics introduced the stunning Atlas 2, its most advanced and human-looking robot yet. Tesla trundled along with Optimus updates, Figure AI arrived with the remarkably bright Figure 01, and 1X Neo Beta‘s uncomfortable human/robot interactions will forever haunt our dreams. The slow pace of humanoid robotic development I’ve witnessed over the last three decades has accelerated partly because some are pairing their robot efforts with advancements in AI. Figure 01, for instance, partnered with OpenAI for some voice-based responses.

When asked, as I sometimes am, when we will have Star Wars-like robots in the home, I usually avoid giving a direct answer because there are too many variables. Everything I know about robotic development tells me the wait will be long.

That, of course, is not the outcome I want. Robots remain at the topic of on my tech wishlist, so I thought that maybe I could find some moral and intellectual support from a few accomplices in the robotics technology race, namely, Google Gemini, OpenAI’s ChatGPT, Siri, and Anthropic’s Claude AI.

A simple robot question

I asked each of them the same simple question: « Will I have a humanoid robot in 2025 »

Siri, even the Apple Intelligence upgraded one, was sadly useless. I kept trying to ask it this simple question, and each time, Siri thought I was in distress and told me, « If you think it could be serious, ask me to call Emergency Services or someone you trust. » I guess Siri assumed I was under robot attack.

I switched to ChatGPT 4o. As I mentioned, OpenAI is already deeply involved in the robot race, working with partners like Figure AI to enable smarter interactions with their existing humanoid robots. Perhaps I shouldn’t have been surprised by ChatGPT‘s largely optimistic response.

ChatGPT told me it was « quite possible that we’ll have advanced humanoid robots. Technology is moving quickly, so you might see some impressive robots by then! Would you be interested in having one? »

I was tempted to tell ChatGPT to « chill » as it asked me what type of robot I wanted. This is a question I found odd since I already told it I wanted a humanoid one.

ChatGPT’s belief in the rise of humanoid robots surprised me, so I challenged it and asked what made it possible for me to have one.

« Well, » it started in its most casual and, yes, human-like voice, « robotics technology has been progressing rapidly, » true, true. « Companies are developing sophisticated robotics with advanced AI, » there it is, « that can perform a variety of tasks. By 2025, it’s likely we’ll see even more practical and affordable humanoid robots. »

That last bit set off a red flag. « Affordable » is not in the 2024 or 2025 humanoid robotics vocabulary. I’m sorry, ChatGPT, but I don’t think you fully understand this market.

Things got more real when I switched to Claude AI and later, Gemini.

Anthropic’s AI threw some cold water on my dream, telling me, « In 2025, it’s unlikely that most people will have a fully functional humanoid robot at home, » adding that « truly versatile humanoid robots are still in early stages of development. »

Instead of Rosey the Robot, Claude AI tells me 2025 will feature:

• More advanced research and industrial robots
• Limited functionality robotic assistants
• More prototype humanoid robots in labs and tech demos

While I appreciated Claude AI’s more clear-headed view, I was also bummed out.

My dreams of a robot out in the backyard playing catch or cooking dinner while it regaled me with fantastic science fiction tales were fading.

Maybe Gemini would lift me up. After all, Google is the land of moonshot projects. Surely, they’d see the humanoid robot glass as at least half full.

Gemini, though, was even more taciturn than Claude AI. Here’s what it told me:

« While there are companies developing humanoid robots, such as Tesla’s Optimus and 1X’s Neo, it’s unlikely that they will be widely available to consumers in 2025. These robots are still in the early stages of development, and it may take several years before they are ready for mass production and commercialization. »

Image 1 of 4

Frustrated mainly because I know Gemini is right, I asked about « the main hold up, » and Gemini had quite a list. What surprised me is that AI sits at the very top of it. Gemini said that building artificial intelligence that can « understand and respond to complex environments [read unpredictable homes] and tasks in a human-like way is still a significant challenge. »

Other challenges include:

• Hardware that can handle all the different kinds of tasks we do in a home.
• Battery tech that can last long enough.
• Cost: It costs not only millions to develop and build these robots, but any humanoid robot you might try to buy in 2025 will cost tens of thousands of dollars.

It’s a harsh dose of reality from at least two of the world’s leading AIs and a lot of wishful thinking from one that should know better.

Despite all this, I look forward to an exciting 2025 in humanoid robotics. Sure, we won’t bring Optimus or Atlas home, but I guarantee we will see some eye-opening updates. That fun will probably start at CES 2025. At last year’s massive tech event, I danced with a humanoid robot. Who knows what they’ll be capable of this year.

You might also like