RFID and BLE bring sensory data to ChatGPT

ChatGPT is an artificial intelligence (AI) chat robot technology. Through its ability to interact with humans and learn from each interaction, ChatGPT is billed as a solution to change everything from student research to mental health treatment. XO2Tech, a technology start-up headquartered in Spain and the United States, said that ChatGPT was enhanced by analyzing RFID transmission to sense the situation around people and products, and by providing information perception data to the system. The company is currently discussing the introduction of solutions into commercial markets such as healthcare, retail inventory management and family personal life management.

ChatGPT, launched commercially in November last year, uses an intelligent function officially known as Generate Pre-Training Converter (GPT), which is based on the OpenAI language model. It focuses on imitating human dialogue, but it can also accomplish things that traditionally belong to human efforts. Writing poetry, music and student essays are just some of the more well-known functions. ChatGPT will remember the prompts received in each conversation, and over time, it can grow with the growth of knowledge.

So what does this have to do with RFID? According to Edward Espinosa, CEO of XO2Tech, an entrepreneur. They have some very exciting synergies, which jointly promote a new type of powerful interactive supply chain and consumer-oriented applications. RFID tags carry out the physical work of data transmission when being queried. Data can be anything from simple ID numbers to sensor information captured based on changes in RF fields. ChatGPT uses AI algorithm to solve problems, so that users can obtain value from these data.

Detection conditions by RF wave

The RFTracker application of XO2Tech collects sensing data by tracking the RF characteristics of passive UHF or NFC RFID or Bluetooth low power consumption (BLE) transmission. When the reader asks for a tag, the tag will respond with its unique ID number. However, the mode of transmission will be affected by the surrounding environment. Espinosa said that RFTracker technology took advantage of this feature. According to Espinosa, RFTracker technology takes advantage of this function.

For example, if someone touches the marked item, the person's body will physically change the transmission of the tag. Then, the software can recognize this change, which will indicate that someone has interacted with the marked items, whether it is a toothbrush, a product on the store shelf or a prescription bottle in the hospital. RFTracker can also detect other conditions, such as whether there is a specific gas that may indicate that the product is starting to deteriorate, or there is a liquid that can identify the leakage or measure the amount of liquid in the bottle, or the vital sign activity of the body, such as blood pressure or heart or respiratory rate.

Users can place labels at strategic locations, such as food packaging, household products or prescription bottles, to track the use of items or products, or the health and behavior of people who interact with them (such as athletes or Alzheimer's disease patients). The user can then determine whether the patient has processed their medication, whether the medication has been removed from the table, or whether they are close to the patient.

Espinosa said that XO2Tech has many patents on this technology and is seeking potential partners to put it into development and test. He pointed out that the emergence of ChatGPT made it more interesting. Developers have used Microsoft Azure to build ChatGPT into a natural voice assistant similar to smart speakers, which can interact with human users in homes or enterprises.

Track health and drug compliance

For example, such an assistant can remind the patient to take medicine. The sensor technology of XO2Tech will enable the ChatGPT system to know whether the patient has taken drugs based on sensory data. The system can identify that the drug is not taken at the expected time, and then send a message to the patient or talk with the patient through the speaker.

The data can also include other sensory information. For example, if mayonnaise is labeled and begins to deteriorate, an alert can be sent through the BLE function on the RFID reader or user's smartphone. The ChatGPT/RF label application of XO2Tech now includes Vital Check, which Espinosa calls the intelligent interconnection packaging solution to provide consumer safety information, such as product freshness and drug compliance.

For example, if the consumer picks up or touches the product in the store or home, the RFID or BLE tag will send a signal, which can be analyzed to determine the physical vital signs of the shopper, including heart rate, respiratory rate, blood pressure, respiratory force, body temperature or physical movement, to detect any adverse reactions to the product. The purpose of important inspection is to send notifications, suggestions or suggestions about the product interaction and vital signs of consumers.

Application for participation in retail business

The company's AIRForce label is dedicated to retail, wholesale, manufacturing and branding. This technology can be built into the label to create the subversive, digital and personalized consumer experience that Espinosa said every product interaction will bring in the entire life cycle of the product, with the goal of reaching, attracting and retaining consumers.

For retail purposes, tracking the time of personal interaction with the product can help the brand or store collect information about the shopper's interests. According to the product detected by the system, the content is then transferred to the personal mobile phone. For example, if they have a store-based or general shopping application, the system can provide suggestions about other products that may be of interest, or invite shoppers to buy products themselves. The content provided can be in the form of personal avatar, video, image, text or voice.

Kitchen-Ai and Ai-Bartender are XO2Tech's first applications based on food and beverage recipes, which use ChatGPT for task guidance. If someone wants to prepare a drink or food menu, the solution can use RFID based sensory data to determine which ingredients are already in the kitchen, which are fresh and which may need to be replaced. Then, users can receive suggestions related to recipes and shopping lists. The application can also detect the user's interaction with the product to determine the next step, or provide suggestions or suggestions.

Espinosa said that XO2Tech also has application potential in other key areas, including supply chain. RFID based sensor labels can be attached to fresh food, infant formula or medicine because these items are stored, transported and sold. The sensor data can then be used to determine the freshness or potential contamination of each product.

Clothing is another product of XO2Tech's patented retail solution. Such a system can be deployed in an individual's home, using RFID readers or mobile phones that support BLE to collect information about each marked clothing. If the user selects a specific shirt, the system can recommend other clothes in the wardrobe to match it. It can also identify unused clothes and suggest donating or recycling them.

Future applications of this technology

Another application is generative marketing using AI and RFID tags. Espinosa said that, in essence, if you touch or interact with a product, it will be detected. Then, through the RFID tag information, the CPU knows what you touch and will provide you with marketing information. This may include the RFID supply chain management label being read when it is within the scope of the reader, and the BLE function that consumers participate in through smart phones.

Dual tags can provide these two functions, as well as sensory data. Espinosa predicts that several applications will take the lead in utilizing ChatGPT and radio-frequency based transmission, such as using Ai Animal Health or Vital Check solutions to analyze health data and provide diagnosis, treatment, products or drugs for the medical industry. He explained that this platform made medical care affordable and convenient. Detect, monitor and analyze the user's physical vital signs, and then conduct generative AI analysis on these data.

This is not for medical advice, Espinosa said. It is for pre-diagnosis or providing information as a medical assistant. This solution can be used by those who seek to cure diseases without using drugs, such as tracking their health status according to diet or supplements.

XO2Tech is continuing discussions with technology companies. Espinosa said that our patent has been put in place, the technology and all applications are there, so we only need to find an interested partner, and we will cooperate with it, which will help us to bring it to the market.

Another company seeking parallel solutions is Wiliot, which has recently obtained a patent for a system and method that is used to determine personal interest in the product in response to battery-free pickup of sensing information.