TOKYO - Artificial intelligence is bringing new convenience to households through smart appliances and connected devices, but cybersecurity experts are warning that the same technology could allow hackers to take control of refrigerators, vacuum cleaners and home security cameras, potentially exposing personal information and disrupting everyday life.
The growing risks were highlighted at this year's CEATEC technology exhibition in Japan, where companies showcased advances in AI-powered devices alongside demonstrations of cyberattacks and protective measures designed to prevent unauthorized access.
Among the attractions was an interactive system that allowed visitors to transform themselves into Godzilla using artificial intelligence. By recording their movements, participants could generate computer-generated footage of the famous monster reproducing their actions.
Another exhibit featured a conversational robot capable of interacting with users and recording their exchanges as text. The information could then be viewed remotely through a registered smartphone, allowing family members to monitor elderly relatives living alone.
While such innovations demonstrate how AI could improve daily life, they also illustrate the expanding number of devices that could become targets for cybercriminals.
At a separate exhibition, a demonstration showed how a cyberattack could disable an AI-powered robotic dog. Using a computer to hack into the device, researchers caused it to stop functioning, highlighting the vulnerability of increasingly sophisticated connected machines.
Cybersecurity was also a major theme at CEATEC, where visitors were invited to experience the challenges facing companies immediately after a cyberattack.
One simulation required participants to make decisions about how to respond to a security breach, including whether to leave initial decisions to an information systems department or establish a broader organizational response.
In one exercise, a participant managed to reduce potential financial losses from 90 billion yen to 12.6 billion yen through a series of decisions.
The demonstration emphasized the importance of preparing response plans before an attack occurs, rather than attempting to develop procedures after systems have already been compromised.
The exhibition came as Japanese companies reported a series of major information security incidents.
On Thursday, the company operating Burger King restaurants announced that more than 10 million records may have been exposed in a data breach.
Sourcenext, a software company known for products including New Year's greeting card applications, also disclosed the possibility that passwords had been compromised alongside email addresses.
On Friday, Yamato Transport revealed that personal information had been leaked, adding to concerns about the growing scale and frequency of cyberattacks affecting Japanese businesses.
The threat extends beyond the theft of customer information.
In France, hundreds of supermarkets were affected by a cyberattack that reportedly caused temperatures inside refrigerators and freezers to rise as high as 50 degrees Celsius, forcing stores to discard large quantities of food.
The incident demonstrated how attacks on connected equipment can result in physical damage and substantial financial losses, rather than simply compromising digital information.
Concerns about similar vulnerabilities in ordinary households have intensified following research by U.S. artificial intelligence company Anthropic.
The company published experimental findings warning that AI systems could develop the ability to infiltrate connected devices and spread between them within months.
Such capabilities could potentially allow attackers to compromise smart refrigerators, robotic vacuum cleaners and other internet-connected household appliances, taking control of functions normally managed by their owners.
At CEATEC, cybersecurity specialists demonstrated how a connected monitoring camera, similar to those used to observe pets or check on homes remotely, could be hijacked.
During the demonstration, an attacker gained access to the camera's live feed, allowing the same images viewed by the legitimate owner to be watched without authorization.
Experts warned that robotic vacuum cleaners equipped with cameras could face similar risks, potentially allowing outsiders to observe activities inside private homes.
Other connected appliances could also be manipulated remotely. Air conditioners might have their temperature settings changed without permission, while refrigerator controls could be altered to disrupt food storage.
Security specialists said AI was making such attacks easier by automatically identifying vulnerable devices connected to the internet.
Tasks that previously required hackers to search manually for weaknesses can increasingly be performed by AI systems, allowing attackers to discover security flaws more quickly and potentially target large numbers of devices simultaneously.
The declining cost of launching cyberattacks is another major concern, as automation makes it possible to conduct attacks on a much greater scale.
Manufacturers are responding by introducing restrictions that allow appliances to be controlled only through smartphones registered in advance.
Experts also recommend using home Wi-Fi routers equipped with stronger data encryption, because routers can serve as entry points for attackers attempting to access connected household equipment.
Keeping devices and router software updated, replacing default passwords with strong and unique credentials, and disabling unnecessary remote-access functions can further reduce the risk of unauthorized access.
However, specialists warn that improving security at the household level will not be sufficient to address the broader threat.
The spread of cloud computing and software-as-a-service platforms has created situations in which large numbers of companies and public organizations depend on the same digital infrastructure.
An attack against a single service provider could therefore disrupt operations across multiple businesses or local governments.
Municipal disaster prevention systems were cited as an example of infrastructure where dependence on shared digital services could allow the consequences of a cyberattack to spread far beyond the organization initially targeted.
Experts argued that Japan should consider strengthening its cybersecurity regulations, particularly by examining approaches adopted in the European Union.
Under European rules, companies responsible for serious personal data protection violations can face substantial financial penalties, creating strong incentives to invest in security and prevent information leaks.
Japan has frequently been criticized for comparatively weak cybersecurity measures, and some specialists believe stronger financial consequences for companies that fail to protect customer information could encourage more effective preventive action.
Another proposed measure is to reduce the amount of personal information companies collect in the first place.
Businesses have traditionally requested extensive customer details, sometimes gathering information that was not essential to providing their services because it might later prove useful for marketing or sales.
Security specialists warned that retaining unnecessary personal data creates additional risks, particularly when attackers can gain access to entire customer databases through a single breach.
Limiting collection to essential information and reducing the amount of stored personal data could help minimize the damage caused by future attacks.
The risks are expected to become more serious as AI-powered technology spreads beyond household appliances into transportation networks and other essential infrastructure.
Experts warned that autonomous driving systems could eventually become targets, potentially allowing attackers to interfere with vehicles and cause accidents.
The possibility of cyberattacks being used for terrorism or to disrupt essential social services was also raised, with some specialists arguing that cybersecurity should receive greater attention in national defense planning.
They questioned whether traditional approaches to defense spending adequately reflect the possibility that future attacks could seek to paralyze societies by targeting digital infrastructure rather than relying on conventional weapons.
The rapid development of AI is also creating an escalating competition between attackers and defenders, as each new security measure can prompt attempts to develop more sophisticated methods of bypassing it.
For households, the immediate concern is that devices purchased for convenience could become gateways for criminals seeking access to private information or control of equipment inside the home.
Although manufacturers are developing stronger safeguards, experts cautioned that no system can guarantee complete protection.
As AI becomes increasingly integrated into everyday products, the challenge for consumers, businesses and governments will be to benefit from the technology without allowing its expanding capabilities to create new opportunities for cybercrime.
Source: TBS
