Did an 18-year old young woman from Texas prove that Quantum Computers aren't exponentially faster?
Actually, she showed that it is possible to create classical versions of Quantum algorithms
Ewin Tang skipped the 4th, 5th, and 6th grades to take classes at a local high school and the University at Arlington at age 10.
Her father is a professor at the University of Texas at Austin, which made it easier for her to start taking classes at the University of Texas when she was 14.
Quantum computers are thought to make certain types of problems exponentially faster, but Tang proved that a “normal” computer could match the speed of a specialized quantum computer if it were given the ability to load data from memory equally quickly.
Quantum computing is not disproven entirely, but Tang does show that the edge some quantum computing algorithm’s claim to achieve is based on having preloaded the data into memory.
QUESTIONS:
How many opportunities do you believe there are to improve efficiency in the economy using the available technology?
Tangentially Related: How long do you project it will take until Quantum computing poses a threat to the encryption used to secure web browsing and banking transactions, dubbed “Q-Day”?
Did you know that Intelligence Agencies and Cybercriminals are downloading your data now, as part of a “Harvest Now, Decrypt Later” strategy?
They are intercepting and storing massive amounts of encrypted internet traffic and banking data today, knowing that in 5 to 10 years, they can feed that archived data into a quantum computer and instantly unlock it. If your data needs to remain secret for the next decade, it is already at risk.
Publications
Quantum principal component analysis only achieves an exponential speedup because of its state preparation assumptions, Ewin Tang, Oct 31 2018
Detail: Gemini Summary:
Ewin Tang is a theoretical computer scientist renowned for “dequantizing” quantum algorithms—creating classical versions that perform nearly as fast as their quantum counterparts. Her work significantly shifted the understanding of where “quantum speedup” actually comes from, proving that many tasks previously thought to require a quantum computer can actually be handled by traditional ones.
The Breakthrough: Recommendation Systems (2018)
As an 18-year-old undergraduate at UT Austin, Tang addressed a major problem in quantum machine learning (QML). At the time, an algorithm by Kerenidis and Prakash was considered one of the best examples of exponential quantum speedup, specifically for recommendation systems (like how Netflix suggests movies).
The Problem: Finding “good” recommendations in a massive dataset.
The Discovery: Tang developed a classical algorithm that used sampling-based techniques to solve the problem in polylogarithmic time.
The Impact: This effectively “killed” the claim of exponential speedup for this specific problem, showing that the advantage wasn’t in the quantum processing itself, but in the way the data was structured and accessed.
Key Research Themes
Tang’s research typically follows a specific framework often referred to as dequantization:
Quantum-Inspired Classical Algorithms: She looks at quantum algorithms that achieve speedups through “state preparation” (a way of accessing data). She then creates a classical analog of this data access—called sampling access—to build classical algorithms that match the quantum performance.
Linear Algebra & Machine Learning: Her work spans various foundational tasks, including:
Principal Component Analysis (PCA): Finding patterns in high-dimensional data.
Matrix Inversion: Solving complex linear equations.
Stochastic Regression: Predicting outcomes in noisy datasets.
Quantum Data & Hamiltonian Learning: More recently, her work has shifted toward understanding how to learn about quantum systems themselves, such as “Hamiltonian learning,” which involves estimating the forces and interactions within a quantum system.
Awards and Current Status (as of 2026)
Tang’s work earned her the 2025 Maryam Mirzakhani New Frontiers Prize for her advances in classical analogs of quantum algorithms and quantum machine learning. After completing her Ph.D. at the University of Washington and a postdoctoral fellowship at UC Berkeley, she is set to begin a professorship at Princeton University in 2026.
Her research continues to be a “reality check” for the quantum computing field, forcing researchers to find truly unique quantum advantages that cannot be replicated by clever classical sampling.
Ewin Tang’s “Dequantization” Explained
This video provides a technical deep dive by Ewin Tang herself into how she dequantizes quantum linear algebra algorithms.
Ewin’s Twitter: @ewintang