AI Highlights — the whole picture — 2026-10-19 (Mon) Morning News

Layoffs in the telecommunications industry had been occurring even before AI was cited as a reason. Now, the same reduction is accompanied by an explanation: AI. In the gambling industry, employees are reporting layoffs due to AI. Beyond employment, AI will be the reason and the means to determine prices, bank transactions, address loneliness, and verify diagnostic imaging. While reading this half-day article, I have one question to keep in mind. Of the events that were said to be ``AI's fault,'' how much did AI actually change, and where did the explanations come from after-thoughts that were added to pre-existing judgments? In the main text, materials for confirming the boundaries are arranged section by section.

AI used as a reason for personnel reduction

Layoffs in the telecommunications industry had been occurring even before AI was cited as a reason. The Sebastian Barros Newsletter points out that telecom companies have been cutting jobs even before AI, and now AI is being used as the reason. Even though the chronology of reductions is the same, only the accompanying explanations have changed. Times of Malta reported that employees in the gambling industry are reporting layoffs due to AI. In two separate industries, the telecommunications industry and the gambling industry, there are stories of reductions due to AI.

When we put these two things side by side, we can see what the reader should check. It is not possible to tell just from the text of the article whether the report that ``AI has reduced the number of people'' refers to the introduction of AI itself or an explanation attached to a pre-existing reduction plan. In the telecommunications industry example, the deciding factor is whether the same reductions occurred in the period before AI. The gambling industry's report is the employee's perception and needs to be verified separately from what the company cited as the reason for the reduction.

Meanwhile, anxiety itself is widespread among working people. KTVU called the fear of losing one's job due to AI "AI grief" and covered how to deal with change. Al Jazeera reports that big tech companies are using fear of AI as a weapon. The stronger the fear, the more important products and investments from companies that use AI will seem. For companies, fear may act to suppress dissent from management decisions by explaining the reason for reductions as AI. This point shows that we need to look for the source of anxiety not only in the workers' actual feelings, but also in the motivations of the people speaking.

You can read in the same context where companies are looking for human resources. Express Computer covered the destination of corporate AI talent through Accenture's competition. If there is a demand for AI talent while there is talk of reductions, then the explanation that it is "AI's fault" may be lumping together different movements in different occupations.

Looking at it this way, when AI is cited as a reason, the strength of the narrative is not necessarily proportional to the certainty of the reality. In the next section, we will see how the same narrative appears in price and financial situations outside of employment.

AI entering life and the market

Based on this material, we can see four situations in which surrounding regulations are not keeping up with the speed at which AI is entering our lives and markets.

In finance and retail, the line between those who use AI and those who receive it is being questioned. trendingtopics.eu reported that US banks are locking out AI agents (www.trendingtopics.eu). Banks do not accept software that operates on behalf of users as a counterparty to transactions. In retail, The Boston Globe reported that the introduction of AI-based pricing is progressing in the United States, raising concerns among consumers. The seller uses AI, and the buyer's AI can be rejected at the entrance. Even with the same technology, there are those who can use it and those who cannot. It is difficult for buyers to ascertain how prices in stores and on screens are determined.

When it comes to content and agriculture, the issue is who receives benefits and information. YouTube Music has introduced three levels of display for AI-generated songs (tech-insider.org). Listeners will be able to distinguish and select the extent to which a song is created by AI. However, it is not clear from the materials how this indication is actually used. In agriculture, AI is transforming the industry, but the results may not reach the farmers who need it most, Bioengineer.org said. There is a difference between the existence of technology and the fact that it reaches the people who need it.

The macro outlook is full of expectations. According to CNBC, Anthropic's Clark told policymakers that rapid advances in AI could significantly increase U.S. GDP (CNBC). This is his opinion and is not a confirmed result. The person speaking is also a person from the company that develops AI. Before using these numbers as a basis for judgment, readers should check who was speaking and to whom.

What all four cases have in common is that the implementation has moved forward, and the rules and how the benefits will be delivered are questioned later. Even if the prospects for growth are bright, the results will not be realized in people's lives unless the individual aspects of banking, price determination, labeling, and delivery to farmers are not in place. In the next section, we will look at how the speed of this introduction widens the distance between what is said and what actually happens.

Loneliness/Emotions/Humanity

AI is also beginning to be used in the areas of loneliness and emotions. The New York Times reported on a scene in which a performer who sang and danced on a Korean TV show was rejected by an AI (The New York Times). This means that AI will play the role of your romantic partner, and will also be on the receiving end of rejection. In Japan, Neyagawa City in Osaka Prefecture has started an initiative to respond to consultations using generative AI to alleviate the loneliness of elderly people living alone (NHK News). The nature of romance in entertainment programs and support for the elderly by local governments are different. However, one thing they all have in common is that AI is infiltrating the places where people talk.

There is also a movement that advocates for humans to be prepared. In Ghana, there was a movement to encourage media professionals to utilize emotional intelligence and AI (3News). How higher education institutions can incorporate AI literacy into their education is also discussed in University World News. What I would like to focus on here is that the skills of usage and the ability to handle emotions are discussed together in the same place. Students are learning not only whether or not they can use AI, but also how to handle their own and other people's emotions when interacting with AI.

However, it is too early to read these stories based solely on the idea that ``AI will destroy human relationships.'' Vietnam.vn points out that fear of the apocalypse existed long before AI (Vietnam.vn). Anxiety itself is not new. AI can be read as a new target for this anxiety. The case of Neyagawa City is the same; loneliness among elderly people living alone has existed since before AI. AI has only been added as a means to deal with loneliness, not the cause of it.

When you organize things this way, you can distinguish between what has changed and what has remained the same. What has changed is that AI has become a candidate for people to talk to and consult with. What hasn't changed is that the source of loneliness and anxiety lies in people's lives. The events of the Korean TV show also show that the opportunities for people to choose whom to direct their feelings to have expanded. Both explanations that blame emotional problems on AI and explanations that say AI will solve them tend to overlook this source.

So, when AI is talked about as an issue of humanity, whose interests and concerns are reflected in that talk? In the next section, we will look at what causes and pretexts for this narrative.

Clinical application verification and safety rules

A meta-analysis of AI interpretation of breast MRI, while acknowledging its high accuracy, urges caution when handling it in clinical practice (Bioengineer.org). The position is that accuracy numbers alone are not a basis for introducing the system, and that confirmation procedures are necessary. Comparisons are also progressing on the method side. A study was published that compared methods for selecting and extracting features in radiomics risk models that predict survival from radiological images (Heliyon). In breast cancer, there are attempts to link MRI-derived imaging phenotypes to molecular subtypes of transcripts and integrate them in an explainable manner (Research Journal of Life Sciences Bioinformatics Pharmaceutical and Chemical Sciences). Research is moving toward showing not only whether the model is correct, but also why it makes such predictions.

The verification stage is also visible in individual cancer research reports. A multimodal MRI deep learning model that predicts bone invasion in meningiomas was subjected to multicenter external validation (Magnetic resonance imaging). For resectable hepatocellular carcinoma, an MRI model incorporating underlying pathology models was used to preoperatively predict clusters of tumor cells surrounding blood vessels. This is a multicenter, retrospective study (Advanced science). Multimodal AI refines EAU risk stratification in prostate cancer radiotherapy (Scientific Reports). For stereotactic radiotherapy for lung cancer brain metastases, THINKERS-X attempted to predict outcomes and exploratory evaluation of dose scenarios using multi-sequence MRI radiomics (clinical & experimental metastasis). A fusion network using multiphoton imaging was demonstrated for predicting lymph node metastasis in pancreatic ductal adenocarcinoma (Biomedical Signal Processing and Control). When readers read such reports, they need to look at the disclaimers as to whether they are externally validated, multicenter, retrospective, or exploratory. Just reading the headline ``AI is highly accurate'' does not tell us whether it is of a level that can be used in the field.

Verification and consensus are also questioned outside of medicine, and the place of inquiry has shifted to the system side. Anthropic is under government sanctions and presented a bill to Congress for the ``AI Safety Act'' (BigGo Finance). The U.S.-China AI dialogue is now in its seventh round, and how to reach an agreement on security is being treated as an issue between the nations. Just as researchers use external verification to verify claims, systems are required to have a verification mechanism that does not rely on business operators' self-reporting.

The more we talk about how AI works, the more we need to look at the verification steps to back up those words.

The situations covered in each section are different: employment, prices and finance, loneliness and emotions, and clinical examination. However, what they all have in common is that if they assume that the explanation is over as soon as the word AI is mentioned, they will misjudge the reality. Reductions in the telecommunications industry were ongoing even before AI. Banks do not accept AI agents as trading partners, and pricing mechanisms are difficult to see from buyers. Even if the accuracy of breast MRI interpretation is high, confirmation procedures are required. Whether AI is the cause or a pretext, clues to confirming the situation lie in the timing, mechanism, and verification records. What readers need to do is look for evidence after receiving the explanation of AI.

Q1: When you introduce AI to your work, how can you distinguish between what has changed and what has remained the same? Q2: When a judgment or explanation is made based on AI, is there a system in place to record the basis in a form that can be verified later? Q3: When citing AI as a reason for organizational decisions, to whom can you explain the relationship with the plans and circumstances that existed before the introduction of AI?

Finally, three questions

- When you incorporate AI into your work, how can you distinguish between what has changed and what has remained the same? Q2: When a judgment or explanation is made based on AI, is there a system in place to record the basis in a form that can be verified later? Q3: When citing AI as a reason for organizational decisions, to whom can you explain the relationship with the plans and circumstances that existed before the introduction of AI? - When a judgment or explanation is made based on AI, is there a system in place to record the basis in a form that can be verified later? Q3: When citing AI as a reason for organizational decisions, to whom can you explain the relationship with the plans and circumstances that existed before the introduction of AI? - When citing AI as a reason for an organizational decision, to whom can you explain the relationship with the plans and circumstances that existed before the introduction of AI?