Washington - A startup led by former "Google X" engineers has launched the "TwinMind" application, touted as a quantum leap in personal artificial intelligence, amidst growing concerns regarding its implications for privacy and digital security.
The app, spearheaded by scientist Daniel George, a former member of a Nobel Prize-winning team, acts as a digital assistant that tracks what users say and do throughout the day, providing precise daily reports to help organize information and overcome forgetfulness in an increasingly fast-paced digital life.
According to the company, TwinMind records context without storing audio, relying on large language models that operate locally on Apple Silicon-powered iPhones, thus allowing for all-day use without significant battery drain. The application offers its services through a free plan and a paid option at $10 per month, with a discounted annual subscription.
TwinMind markets itself as a "non-forgetting memory," claiming it addresses what the human brain tends to forget—90% of memories within a week. Users can delve deeply into their daily logs, generate messages and reports, and receive advanced study support by summarizing lectures and creating flashcards. Additionally, the app, also available as a Chrome extension, allows users to interact with tabs, write texts, and input data on any site, making it akin to a "digital butler" present at all times.
However, the app faces a wave of criticism, especially concerning digital privacy. The idea of a continually listening application, even with the company's assurance that it does not store audio, raises alarms among digital rights advocates. The app's privacy policy indicates that user data may be shared with external service providers, including cloud storage companies, analytics providers, and security services, increasing the risk of sensitive data breaches or exploitation.
A recent study published on arXiv confirmed that "digital twin mind" applications could become targets for cyberattacks due to their detailed knowledge of user behavior and devices. Research from Columbia University has also shown that listening AI systems can be deceived by AI tools specifically designed to disrupt information.
The developers of TwinMind assert that the app marks the beginning of a new era in personal AI, with Daniel George predicting that users will fully rely on such tools in the future, stating: "Once they try it, they won't stop using it." However, privacy and cybersecurity experts are calling for regulation of this emerging sector to ensure these applications do not turn into digital surveillance tools disguised as personal assistants.
Some are advocating for legislation that mandates complete transparency on how data is collected and processed, granting users genuine control over their information. As reliance on AI applications increases, users find themselves facing a challenging equation: reaping the tremendous benefits of this technology while risking exposure to unforeseen dangers.
Does TwinMind represent the future of digital assistance that will transform our daily lives, or is it merely another step toward deepening the privacy crisis in the digital age? This question remains open as these technologies evolve and companies race to occupy the "twin mind" space in every user's pocket worldwide.
