Breaking
Hong Kong’s 5-year plan consultation has sparked valuable discussionsClop builds custom web shell for Windchill servers in data theft campaignTop Chef turns to tears as host Kristen Kish redefines the roleMicrosoft Copilot Personal Flaws Could Let One Click Exfiltrate Data From Connected AppsMLflow flaw exploited to steal cloud credentialsESMA consults on reporting framework for clearing activity at recognised third-country CCPsECB collateral surprises cut bank risk and narrow bond spreadsFlores power restored after earthquake as all 1.144m customers back onlineAI firms undercount personal use of their tools as scepticism grows over dataCursor launches rival code-hosting platform Origin as GitHub struggles with reliabilityOil prices rise to three-week high as Strait of Hormuz threat lingersJustice Department probes Andreessen Horowitz over board seats at Databricks and FivetranPeople are still unsure how AI is being used in practiceEuropean stocks fall to two-week low as bond yields rise and Middle East tensions simmerRFC 9234 adoption lags in Malaysia as route leaks threaten banks and telcosPublic Service DG Wan Ahmad Dahlan named MRT Corp chairmanPuncak Niaga sells two Selangor plots for RM237.15m to cut debtAWS to open permanent Builder Lofts in Berlin, Hyderabad and São PauloRTS Bukit Chagar station to link directly to KTM, buses and taxisThreat Brief: Mitigating Large-Scale Credential Attacks (Updated Aug. 18)Hong Kong’s 5-year plan consultation has sparked valuable discussionsClop builds custom web shell for Windchill servers in data theft campaignTop Chef turns to tears as host Kristen Kish redefines the roleMicrosoft Copilot Personal Flaws Could Let One Click Exfiltrate Data From Connected AppsMLflow flaw exploited to steal cloud credentialsESMA consults on reporting framework for clearing activity at recognised third-country CCPsECB collateral surprises cut bank risk and narrow bond spreadsFlores power restored after earthquake as all 1.144m customers back onlineAI firms undercount personal use of their tools as scepticism grows over dataCursor launches rival code-hosting platform Origin as GitHub struggles with reliabilityOil prices rise to three-week high as Strait of Hormuz threat lingersJustice Department probes Andreessen Horowitz over board seats at Databricks and FivetranPeople are still unsure how AI is being used in practiceEuropean stocks fall to two-week low as bond yields rise and Middle East tensions simmerRFC 9234 adoption lags in Malaysia as route leaks threaten banks and telcosPublic Service DG Wan Ahmad Dahlan named MRT Corp chairmanPuncak Niaga sells two Selangor plots for RM237.15m to cut debtAWS to open permanent Builder Lofts in Berlin, Hyderabad and São PauloRTS Bukit Chagar station to link directly to KTM, buses and taxisThreat Brief: Mitigating Large-Scale Credential Attacks (Updated Aug. 18)
Economy

AI firms undercount personal use of their tools as scepticism grows over data

An independent research project has found that people use artificial-intelligence tools far more often for sensitive personal tasks than major providers report, raising fresh questions about how AI companies shape public understanding of their products.

Source: MIT Technology Review · August 18, 2026 at 10:31 PM · AI-assisted report

Single-source

KUALA LUMPUR, 19 AUGUST 2026 —

Listen to this article

DomainFork Audio · read aloud

AI Usage Patterns Revealed in Independent Study, While Flock Safety Adjusts Surveillance Model

Market Impact

KUALA LUMPUR, Aug 18 — A new research initiative, the AI Observatory, has shed light on how people in the United States are using artificial intelligence tools, revealing patterns that differ from those reported by major AI developers such as OpenAI and Anthropic. The study, which aims to provide an independent analysis of AI usage, found that personal applications of AI—such as roleplay and social interactions—are far more prevalent than what corporate reports suggest.

Meanwhile, Flock Safety, a U.S.-based police technology firm, has announced design changes to its surveillance platform to curb potential misuse by law enforcement.

The AI Observatory’s findings challenge the transparency of data released by leading AI companies, which have historically focused on professional and educational uses of their tools. According to the study, users turn to Anthropic’s models primarily for coding tasks, while Google’s Gemini is favored for social and roleplay scenarios. ChatGPT, developed by OpenAI, remains the most commonly used platform for homework assistance.

These insights were derived from an analysis of real-world usage patterns, offering a more nuanced view of AI adoption beyond corporate narratives.

Flock Safety operates a network of approximately 120,000 automatic license plate readers (ALPRs) across the U.S., a system designed to assist law enforcement in crime-solving. The company has now introduced updates to its platform aimed at preventing officers from using the technology for unauthorized purposes, such as stalking. While proponents argue that such surveillance tools have helped capture suspects in serious crimes, critics question the broader implications for civil liberties.

The design choices behind Flock’s system—including what data is collected, who can access it, and how long it is retained—define the balance between security and privacy.

The AI Observatory’s research underscores the need for independent scrutiny of AI usage, as corporate reports often omit sensitive or personal applications. For instance, the study highlights that roleplay and social interactions account for a substantial portion of AI engagement, areas that are rarely emphasized in official disclosures. This discrepancy raises concerns about the completeness of data provided by AI developers, particularly as these tools become more integrated into daily life.

In Malaysia and Southeast Asia, where AI adoption is growing rapidly, the findings could influence how governments and businesses approach the regulation and deployment of AI technologies. The region has seen increasing investment in AI-driven solutions, from smart city initiatives to financial services automation. However, the lack of independent data on AI usage in these markets means that policymakers may need to look beyond corporate reports to fully understand the societal impact.

Stakeholders in Malaysia’s tech ecosystem have expressed mixed reactions to the AI Observatory’s revelations. Industry analysts note that while the study provides valuable insights, its focus on the U.S. market may not fully reflect regional trends. “AI usage in Malaysia is still evolving, and we’re seeing more adoption in sectors like healthcare and education,” said a spokesperson for Malaysia Digital Economy Corporation (MDEC).

“However, the emphasis on personal and social uses of AI aligns with global patterns, particularly among younger users.”

The Flock Safety case also resonates in Malaysia, where discussions about surveillance and data privacy are gaining traction. The government has introduced the Personal Data Protection Act (PDPA) to regulate the collection and use of personal data, but concerns persist about the potential for misuse by both public and private entities. “The balance between security and civil liberties is a global challenge,” said a cybersecurity expert based in Kuala Lumpur.

“Any system that collects large-scale data must have strict safeguards to prevent abuse.”

Looking ahead, the AI Observatory plans to expand its research to include more regions and user demographics, providing a broader picture of AI adoption worldwide. For Flock Safety, the recent updates to its platform may serve as a model for other surveillance technology providers seeking to align their operations with ethical and legal standards.

In Malaysia, where digital transformation is a key economic priority, the findings could prompt further scrutiny of AI tools and their societal impact.

As AI continues to permeate various aspects of life, the demand for transparency and independent oversight is likely to grow. The AI Observatory’s work represents a step toward filling the gaps left by corporate reports, offering a more comprehensive view of how these technologies are truly being used. For policymakers, businesses, and users alike, understanding these patterns will be in shaping a future where AI serves the public interest without compromising fundamental rights.

Additional developments in global AI and technology this week include Nvidia’s $105 billion commitment to OpenAI’s Ohio data center, Tesla’s upcoming launch of robotaxi services in Austin, and China’s push to shape AI datasets with “mainstream Chinese values.” Details not yet available on the regional impact of these developments in Malaysia.

Related: Google · Kuala Lumpur

Reporting based on MIT Technology Review. Figures and claims are subject to revision as the story develops. DomainFork publishes editorial context, not investment advice — see our editorial standards.