Physical AI: How NFC Tags Can Make Everyday Products Intelligent

Artificial intelligence has, until now, largely existed behind a screen. We type a question into a website, talk to an app or upload information to an online service.

Physical AI changes that relationship. Instead of starting with a screen, the interaction starts with a physical object.

A customer taps a jacket with their phone and asks how it should be washed. A visitor taps a label beside a tree and asks when it flowers. An engineer taps a machine component and requests the correct installation instructions. The object becomes the starting point for an intelligent, relevant and potentially unique conversation.

NFC tags provide the connection between the physical object and the AI system. More importantly, authentication NFC tags can make that connection secure and specific to an individual product.

What is physical AI ?

The term "physical AI" is often used to describe AI systems that understand or operate in the physical world. This can include robotics, autonomous vehicles and intelligent machines.

However, there is another, more immediate application: enabling ordinary physical objects to provide intelligent information and interaction.

The object itself does not need to contain a processor, battery or AI model. Instead, an embedded NFC tag gives it a secure digital identity. When the tag is scanned with a mobile phone, the user can access an AI-driven experience connected to that particular object.

This creates a simple interaction:

1. The user taps the object with an NFC-enabled phone.
2. The NFC tag identifies or authenticates the object.
3. The user is taken to a relevant digital experience.
4. AI uses the available information to answer questions or provide guidance.

No dedicated app is necessarily required. From the user’s perspective, the process can be as simple as tapping the product and asking a question.

Giving every object its own information

A useful physical AI experience needs context. A general AI system may understand what a waterproof jacket is, but it will not automatically know the materials, coatings, manufacturing date or care requirements of the particular jacket being scanned.

Seritag’s ixkio NFC Tag Authentication and Management platform addresses this by organising information in a hierarchy.

At the top level is company information. This might include general brand details, customer-support policies or information relevant to every product made by the organisation.

Below that is product-level information. This can contain instructions, technical specifications, care guidance and other content relating to a particular model or product range.

At the individual tag level is information associated with one specific item. This might include its serial number, date of manufacture, warranty status, service history, batch information or ownership details.

This structure creates what can be considered an information silo around each NFC tag. The AI experience is not simply responding with information collected from the wider internet. It can respond using controlled information relevant to the company, the product and, where required, the exact physical item.

A single tap can therefore start a highly specific interaction.

A single tag - multiple languages

Ixkio's powerful AI platform can even understand language. While the original instructions can be written in English, a person scanning the tag and asking a question in Spanish will be given the answer in Spanish. 

Even more, it can understand and help if the user needs more help. While the instructions can be given quickly, someone asking for more help can be given a simpler breakdown of information. All from the same original text. 

From static webpages to intelligent conversations

NFC tags have been used for many years to connect products to webpages. A customer taps a tag and opens a product page, instruction manual or registration form.

Physical AI develops this concept further.

Instead of requiring the customer to search through menus, PDFs and support pages, the NFC interaction can allow them to ask a direct question:

• How do I clean this product ?
• Is this component compatible with my machine ?
• Is my warranty still valid ?
• How do I assemble this item ?
• What material is this made from ?
• Where can I order a replacement part ?
• Has this product been serviced ?
• How should this item be recycled ?

The answers can be generated from information already associated with the product and tag. This can make support easier for the user while allowing the brand to control the underlying source information.

The result is not just a smart product page. It is a conversation with the physical object.

Why NFC authentication matters

Standard NFC tags can give an object a digital identity, but standard tag data can potentially be copied. For basic marketing or information applications, this may not be a significant concern. For object-specific AI interactions, however, identity and authenticity can become essential.

Authentication NFC tags generate dynamic cryptographic data when scanned. This data can be checked by an authentication server to confirm that the interaction originated from a genuine tag rather than a straightforward copy. Seritag’s ixkio platform can provide the tag-management and authentication layer required to deploy these tags at scale.

This gives authentication NFC tags an important advantage for physical AI: the digital interaction can be securely connected to the physical object.

Without physical access to the genuine product and its embedded authentication tag, it becomes substantially more difficult to reproduce the valid interaction. A copied link or photographed code is not equivalent to possessing and scanning the original authentication tag.

This is particularly important when the information is unique to an individual item. Warranty details, ownership records, maintenance history or serial-number-specific instructions should not be accessible through an easily copied identifier.

The way the tag is integrated into the product also matters. An authentication tag confirms the tag’s identity, so it should be embedded or attached in a way that makes removal, transfer or substitution difficult. Depending on the application, authentication can also be combined with tamper-detection technology.

Clothing that explains how to care for itself

Clothing provides a straightforward example of how physical AI could work.

Care labels are limited in size and often depend on symbols that users may not understand. Printed instructions also remain fixed, even if guidance, repair services or recycling options change.

An NFC tag embedded into a garment could provide access to detailed, up-to-date care information. Instead of interpreting a row of symbols, the owner could tap the garment and ask:

"Can I wash this at 40 degrees ?"

"Can this jacket go in a tumble dryer ?"

"How do I remove a stain without damaging the waterproof coating ?"

The system would know which garment was scanned and could answer using information for that specific product. At an individual-item level, it might also provide purchase date, warranty coverage, repair history or information about the materials used in that particular production batch.

Later in the product’s life, the same interaction could help with repair, resale, authentication and recycling.

Allowing visitors to interact with plants and trees

Seritag is already working with botanical gardens to supply NFC tags that allow visitors to interact with plants and trees using their phones.

A conventional plant label has very limited space. It might display the species name, origin and a short description. An NFC-enabled physical AI experience can offer considerably more.

A visitor could tap the label and ask:

"Is this tree native to the UK ?"

"How old can this species live ?"

"Why is it losing its leaves ?"

"When does it flower ?”

"Is it toxic to dogs ?"

The experience can respond using information associated with that plant, species, collection or location. It can also adapt the level of detail to the visitor, providing a simple answer for a child or a more technical explanation for a horticultural student.

The amount of information available can run into thousands of words - far more than you could ever print on a label. And importantly, far easier to access. 

The future of intelligent products

Physical AI could change the role of product documentation and customer support.

Furniture could provide assembly guidance based on the exact model. Industrial parts could retrieve technical data and service records. Medical equipment could display controlled operating information. Consumer electronics could offer setup support without requiring the user to find the correct manual. Museum objects could answer questions according to the visitor’s language and level of knowledge.

The same embedded tag could continue to support the product throughout its life. Information could be updated without replacing the tag, while the experience could change according to whether the person scanning it is a customer, technician, retailer or authorised member of staff.

This is where NFC tag management becomes particularly important. Deploying a few tags is relatively simple. Managing thousands or millions of products, product groups and individual serialised objects requires a platform capable of controlling the information and interactions at each level.

Connecting AI to the real world

NFC tags provide a simple bridge between physical objects and digital systems. AI makes that bridge conversational and responsive. Authentication technology helps ensure the interaction remains connected to the genuine physical item.

Together, these technologies allow almost any product or object to become a source of intelligent information.

The next stage of connected products may not involve adding screens, batteries or complicated electronics to everything we own. It may be as simple as embedding a secure NFC tag and allowing each object to explain itself.