The Tamesis
January 5, 2026
Tokyo – An investigation by a cybersecurity company found that certain comments were unusually spread on social media during the Senate election last July.
About 9,400 social media accounts were found to have signs of bot behavior, such as mass retweeting to amplify specific claims. Analysis of a sample of these accounts showed that such claims were spread rapidly by about half of the accounts in July.
The Japanese company Nexus Intelligence compiled a report after the investigation, specifically covering the period from the official start of the campaign on July 3 to the voting and counting of votes on July 20.
Using social media analytics tools, the company detected about 9,400 accounts exhibiting bot-like behavior, such as spreading posts by repeatedly retweeting them within a short period of time. The company also examined the activity of 170 social media accounts from January 2025 to July of that year, which were deemed more likely to be bots based on characteristics such as account information and creation dates.
The study found that during the July election, certain statements were widely spread through 77 accounts (45% of 170 accounts).
Some of these accounts primarily spread criticism of specific politicians or the advocacy of specific political parties. The spread of such posts undoubtedly leads to increased exposure to specific ideologies, the report said.
Regarding the upper house election, a senior Japanese government official said that “our country has become one of the targets of influence operations” and foreign forces interfered in the election through social media.
However, the chief analyst at Japan Nexus Intelligence said: “It is difficult to draw a clear conclusion. We cannot be sure whether there is interference.”
“A time has come when election outcomes can be significantly influenced by the ability to build passionate communities and create compelling narratives that resonate with people,” the report said.
“Even if the post and its dissemination are unscrupulous in nature, it will be difficult for viewers to determine the intent of the informant,” the analyst said. “People should be consciously skeptical when seeing information on social media.”
