Last month, the Nubia NaviX Ultra officially hit the market. With a starting price of 5,999 yuan (or 5,499 yuan after government subsidies), Nubia announced that its first-day sales surpassed 100 million yuan in just one second.

This mass-produced device, known as the 'Doubao Smartphone II,' is designed to 'help you get things done.' For instance, a user can simply say, 'Find me the cheapest power bank,' and the phone will autonomously open apps, compare prices, and place the order. Similarly, saying, 'Order me some fried chicken,' prompts the phone to launch food delivery apps and suggest the best options based on user preferences.

However, many consumers discovered after purchasing the phone that while the AI assistant can open apps like WeChat, JD.com, and Meituan, it cannot complete in-app operations—at least not within the first thirty days. This limitation arises because Doubao introduced the Screen Automation Operation Statement Protocol (SAEP), which allows third-party apps to declare whether they accept GUI agent operations, with a 30-day public notice period. If mainstream apps explicitly refuse after 30 days, this AI smartphone may indeed become, as netizens joke, a 'ByteDance-exclusive machine.'

From its first to its second generation, ByteDance seems determined to continue experimenting with AI smartphones in its quest to define the next era of mobile interaction. However, smartphones are ultimately tools for human use. Beyond technical challenges, do users truly need an AI smartphone that handles tasks and boosts efficiency? This is a crucial question that major manufacturers must consider.

Entertainment Takes Center Stage

Since the original iPhone's release in 2007, smartphones have evolved for nearly two decades, becoming increasingly sophisticated in both manufacturing and functionality. Our reliance on them has also grown stronger. But what do most people actually use their phones for?

The data is clear: entertainment. According to the China Network Audio-Visual Development Research Report (2026), the average daily usage time per network audio-visual user is 201 minutes, or approximately 3 hours and 21 minutes. Among these activities, short videos are the only segment maintaining 'dual growth,' ranking first in usage time.

How much do Chinese people love watching short videos? A report from QuestMobile in October revealed that among post-2000 users' mobile internet habits, 'watching videos' accounted for 37.1% of their time, making it the absolute mainstream. On average, they spend about 2.7 hours daily on mobile videos. The report also highlighted that beyond 'watching videos,' mobile socializing (25.2%) and mobile gaming (9.7%) constitute the 'top three hobbies' for post-2000 smartphone users.

A notable shift in user behavior is Douyin surpassing WeChat in usage time, undoubtedly reflecting changing public demands. People no longer seek more 'efficient tools' but rather more 'emotional outlets.' This drives users to prefer instant, effortless, and low-cost forms of happiness.

This raises a question: when we watch short videos, play games, or chat with friends, do we need AI to enhance efficiency?

The answer is likely no. Whether watching short videos, streaming dramas, socializing, or listening to music, the commonly used apps on smartphones have undergone countless iterations, allowing users to enjoy happiness with near-foolproof operations. Short videos, in particular, only require constant swiping, eliminating even the need for searching.

In other words, most functions of modern smartphones have already reached peak efficiency.

Of course, beyond leisure and entertainment, we also use our phones for shopping, ordering food, traveling, and other activities. While AI may streamline the selection process during online shopping, reducing time costs, many people actually derive pleasure from browsing and comparing products, making the unboxing moment even more satisfying. The same applies to food delivery.

In fact, can we think of anything we frequently do on our phones that cannot be accomplished without AI or could not be done better manually? Perhaps only some work-related tasks, like writing weekly reports, editing images, or opening PowerPoint presentations. However, these needs are infrequent, and many people do not even use their phones for work at all.

AI taking over smartphones to handle tasks for users essentially aims to boost efficiency and save time and effort. Yet, users who are already heavily reliant on their phones and spend excessive time on them are precisely those who are least short on time.

Should Everything Be Delegated to AI?

A couple of years ago, the term 'brain rot' went viral online, referring to the decline in mental and intellectual states due to excessive exposure to low-quality digital information, especially on social media. This term gained global attention after Oxford University Press selected it as the Word of the Year, sparking widespread debate.

The rise of this concept reflects public-level doubts about the fragmented content controlled by algorithms, triggering widespread anxiety about cognitive decline. Online, we see various critiques of excessive smartphone and social media use, with many people realizing they spend too much time on screens.

Last year, Psychological Bulletin published a study examining the relationship between short video consumption, cognitive ability, and mental health. Researchers reviewed 71 existing studies involving 98,299 participants and conducted a systematic review and meta-analysis. The findings revealed a moderate negative correlation (r = -0.34) between short video usage and cognitive function, meaning more frequent viewing often correlates with poorer cognitive performance. Specifically, attention and inhibitory control were the most affected cognitive abilities.

Frequent short video consumption can impact cognitive ability. What happens when we delegate tasks to AI, let AI assistants complete tasks for us, or turn to AI for help whenever we encounter problems?

A reasonable speculation is that we may become lazier thinkers.

Some time ago, a paper titled 'The Learning Penalty of Generative Artificial Intelligence: Evidence from Secondary Education in China' tracked 26,811 junior and senior high school students in a county in central China. It found that while using generative AI to outsource homework temporarily improved students' assignment scores by about 18%, their monthly exam scores declined by an average of 20% within six months.

The paper even reported on high school and college entrance exam results. Among students who used AI for at least two years, their high school entrance exam scores dropped by approximately 24%, and their college entrance exam scores fell by about 18%.

For students, efficiency is merely a means to an end; the ultimate goal is to enhance learning ability. Using AI tools provides them with results but replaces the thinking process. While adults use AI tools to boost efficiency and free themselves from repetitive tasks, excessive reliance on AI's convenience and efficiency may lead to similar outcomes as with students.

Because this technological revolution is driven by generative AI, a tool that enables us to outsource cognition and thinking more than ever before. With a simple input, we can instantly obtain answers to almost any question, write essays or code, create art, or design works. This represents a 'dissolution' of the human thinking process, perfectly catering to human laziness.

AI smartphones are no exception. In the future, a single phrase may suffice to await results. However, by then, many people will not transition from meaningless labor to valuable pursuits but will instead become increasingly lazy.

Sense of Loss of Control

In the past two years, emotional consumption has surged, and the 'happiness economy' remains popular. Netizens coin terms like 'finding oneself in old age,' 'prefrontal cortex-friendly,' and 'life's odyssey,' all reflecting a clear and widespread psychological need: as the external world becomes increasingly uncertain, young people crave a sense of certainty.

For instance, they may claim to be broke yet willingly spend tens of yuan on a cup of coffee. They advocate for consumption downgrading but will pay for small toys, fragrances, performances, or ceremonial meals.

This psychological reflection is evident in consumer behavior and, to some extent, represents many people's attitude toward their current lives.

The opposite of certainty is a sense of loss of control. For AI smartphones to handle tasks, AI assistants must replace users in navigating various apps. But can users truly trust AI to manage a private device containing vast amounts of personal information? This is one question. Another is that long-term reliance on AI may foster a sense of loss of control: it is no longer 'I' controlling my phone but AI.

For example, in the future, a user might say, 'Book me a hotel,' and quickly receive a confirmation. However, users often only see the conclusion without knowing what AI filtered out or why it chose a particular option.

This represents a shock to past consumption habits, easily evoking discomfort or resistance. More critically, during AI's evaluation and selection process, users are unaware of which products were excluded, the criteria used for ranking, or whether recommended slots involve commercial partnerships.

This lack of transparency naturally breeds distrust.

Of course, when we watch short videos, we are also unaware of the algorithms' recommendation mechanisms, with platforms controlling ranking weights and traffic rules. However, this opacity is tolerable because algorithms only influence a list of 'what you see'—you can like it or swipe it away. But once AI assistants take over smartphones, a series of 'executed actions' accompany each command. If users are unsatisfied with the results, they must re-enter commands and regenerate outcomes.

If this process repeats several times, users may begin to doubt or resent the AI assistant's efficiency.

Moreover, while watching short videos is free, using AI to filter information often aims to facilitate consumption, involving spending. Users will inevitably remain stringent about the experience and efficiency in such cases.

Furthermore, a capable AI assistant within a smartphone must understand users' information comprehensively. Leaving aside whether major companies are willing to grant intelligent agents access, such an 'entity' that knows users better than they know themselves may represent not convenience but fear for many.

We cannot deny that, amid the sluggish smartphone market, an AI smartphone capable of handling tasks represents a bold exploration to break through innovation bottlenecks. However, amid intense competition among tech giants, any innovation regarding smartphones—humans' 'second selves'—should first ask whether users need it.