I haven’t seen Emiglio in years – probably since the mid-’90s when I would see him on TV and sigh sadly. But then one night a few months ago, in a bar in Porto, suddenly he was there, in the flesh – or should I say, the plastic.

Emiglio, you see, is a robot. A knee-high butler with a bulbous white head, cartoon grin and red glowing eyes. I spent my childhood wishing he would bring me fun drinks on his little tray while I sat on the couch, glued to back-to-back episodes of Animaniacs. In retrospect, my desire for Emiglio is probably why I committed myself to a career writing about technology.

Read more: CNET Picks the Best of the CES 2026 Awards

Like many of the fantasies people harbor about robots, this is a bad idea. Who made the fun little drinks and balanced them on Emiglio’s tray? How did he know where to take them? And who, in the end, can help him navigate the two rooms from the kitchen to me?

Many of the issues that prevent Emiglio from being a truly useful butler are the same limitations that robots in the world still struggle with today. As much as I hate to admit it, Emiglio is little more than a glorified remote-control car with a face, requiring human assistance to do almost anything.

The same is true of Neo, the humanoid home helper robot that went viral in late October but still requires teleoperation by a human. The two robots are separated by more than 30 years, but their real-world usefulness and ability to operate autonomously feels equally disappointing.

A question I ask myself every year when I return to CES — the giant tech trade show that the CNET team decamps to every January — is whether the many robots I met there will finally prove worthy of a place in our homes.

Overcoming the AI ​​bottleneck (and why VLA matters)

“The main barrier between us and truly useful home robots is AI,” noted computer scientist Ben Goertzel said when I sat down with him at the tech-focused Web Summit in Lisbon last month. The physical capabilities of the robots improved greatly between Emiglio and Neo. What keeps Neo, and other home robots, in the end are the smarts.

The breakthroughs in AI that we have witnessed in the last few years have paved the way for change in this regard. The great language models created by companies like OpenAI, Google and Anthropic enable us to have more nuanced conversations with our technology, which is particularly compelling in the case of emotional or companion robots.

It is possible that this year at CES, we will see a company that integrates more advanced AI into a robot concept that it has already shown, said Ben Wood, chief analyst at CCS Insight. He suggested that Samsung could build on its rolling projector robot, Ballie, by working closely with Google – as it has already done with phones – to create a next-generation version with Gemini, for example.

“Generative AI allows for more natural language interaction with smart devices,” Wood said, but it’s the same whether you’re talking to a smart speaker, a robot vacuum cleaner or a humanoid.

What is more useful for robots than LLMs is the development of vision-language-action models, which, as the name suggests, is a type of AI that allows for the input of a combination of images and words, and the output of actions. For robots navigating a physical space, this combination is essential and will really set them apart from other AI-equipped devices.

“More advanced models for robotics combined with more capable and more deeply integrated generative AI will see some more intelligent use cases come in either high-end humanoid science fiction robots or some of the more practical robots,” Wood said.

There are debates whether a breakthrough for robots should unlock AGI – a hypothetical level of AI superintelligence. Goertzel, who works in superintelligence and robotics, doesn’t think this is the case. VLAs are getting so good, we don’t need AGI to make decent home robots, he told me.

Looking beyond the humanoid form

The big challenge for home robots is that every home is different.

Places like hotels, schools and hospitals have enough similarities that robot navigation can be more or less standardized. But developing robots for business and industry, where they will perform repetitive tasks in predictable environments, is completely different from training a robot that you can put in different layouts of houses.

Still, some are trying. The Sunday Robotics team, based in California, trained its humanoid robot, Memo, with data provided by families across the US who use high-tech gloves to capture the intricate movements of their hands as they do household chores. It’s an ambitious approach to preparing robots for family life, and if Sunday Robotics can keep up with its desired timeline, it could be one of the first companies to deploy non-teleoperated humanoids in people’s homes.

But for some, there is a real question mark as to whether we should target human domestic helpers.

“When I think about everyday things, like housework, a humanoid body doesn’t feel right,” Goertzel said. “When I think about our kitchen at home, my wife needs me to reach high things, and I don’t like crawling on the ground to fix low things, because we’re a little different in height. Why make that a problem for a robot?”

Instead of equipping an expensive humanoid at home with the same obstacles we struggle with as humans, he envisions a networked system of smaller, more practical bots that can interact and are designed to excel at specific tasks.

There are opportunities for established tech companies to jump in here, whether that’s Samsung with Ballie and Apple with mysterious rumored home robot plans, or companies like Qualcomm, which is at CES and may talk about its own robotics plans at the show.

Qualcomm already makes chips for cars (which are close relatives of robots, especially in their autonomous form) and a whole range of small consumer electronics that maximize power while providing long battery life and AI capabilities. At the Web Summit, CEO Cristiano Amon told me that he sees robotics as “an incredible opportunity.”

“We’re excited,” he said. “Even from business to consumer, I think the type of silicon we’re making for phones and for (edge ​​computing) is the perfect silicon for robots.”

The practical robot: Task-specific bots

Many of us are already starting to invest in smaller, task-specific bots by purchasing robot vacuums, mops and lawnmowers — an established category that will continue to grow on the back of CES 2026.

“There’s been a whole avalanche of robotic vacuum cleaners and mops and robotic lawn mowers,” Wood said. However, he says, you need to have “the right kind of housing” for them to work well.

Still, CCS Insight research suggests that 15% of households across the US, UK, Spain, France and Germany intend to buy a robot vacuum cleaner by 2026. They’re not the coolest or cutest robots, but they win when it comes to their usefulness and what people want to spend their money on.

Like a humanoid home robot? “Honestly, it’s been years,” Wood said. “People love the idea of ​​it, but it’s a long, long, long way from being something that people have in their homes or want.”

The urgent question of privacy and safety

His prediction was echoed by Boston Dynamics CEO Robert Playter, who spoke Euronews at the Web Summit that he believes that robots will not be in our homes for at least five to 10 years. (This from a man whose company made the acrobatically inclined Atlas humanoid and the fearsome Big Dog, both of which were sniffed out by the US military.)

There are many reasons you might not want the futuristic robot in your home, from practical considerations like space and utility, to larger concerns about privacy, safety and cost (the Neo is priced at $20,000, and Sunday Robotics told me the Memo will be a “high-end” product).

In November, robotics researchers at Carnegie Mellon University published a paper stating that popular AI models are not yet ready to power robots because of issues ranging from bias and discrimination to unsafe physical behavior.

The study, which analyzed ChatGPT, Gemini, Copilot and HuggingChat, found that most models are willing to approve orders that compromise people’s mobility aids, allow a robot to brandish a knife, take unwanted photos or steal credit card information.

“If an AI system is to direct a robot that interacts with vulnerable people, it should be held to standards at least as high as for a new medical device or medicine,” said co-author Rumaisa Azeem, a research assistant at the Civic and Responsible AI Lab at King’s College London. There is an urgent need, he said, for comprehensive and routine risk assessments before these AI models are put into robots.

Just a week after the study was published, Figure AI began sued by a whistleblowerwho warned that the company’s humanoid robots could “break a man’s skull.”

As compelling as they are to observe, robots like Elon Musk’s Tesla Optimus and Boston Dynamics’ stable of bots can be a little scary. But when these types of robots are shown at CES this week, the focus will likely be on the skills they can bring to industrial settings rather than the home.

Robotics will be a “mega trend” at the show, according to Wood, but it remains to be seen if any of the robots on display will rise above robot vacuums, pool cleaners or cute but one-note companion devices to become a true household necessity.

As for me, I still dream of owning Emiglio — even more so after finally meeting him. It’s tempting to go on eBay to see if anyone is selling this ’90s robot toy I built in my career, but it’s probably better to wait for the real thing. For now, I will continue to hope that the truly capable (and safe) robot butler of my childhood dreams will one day become a reality.