TL;DR
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
A paper by David Turturean of Physical Superintelligence says it proves the Aaronson-Ambainis conjecture, an open question about quantum algorithms and randomly generated inputs. The result limits quantum advantages in the number of input queries for yes-or-no problems; it does not set a general limit on total computing time or establish which quantum applications will be useful. PSI says an AI-assisted system found the proof, which was formalized in Lean and checked by its kernel.
A paper by David Turturean of Physical Superintelligence (PSI) reports a proof of the Aaronson-Ambainis conjecture, a question dating to 2008 about whether quantum computers can gain exponential advantages on most randomly generated inputs. The result applies to the number of times an algorithm reads its input, not its total running time; PSI says an AI-assisted research system found the proof, which was formalized in Lean and checked by the proof assistant’s kernel.
The theorem concerns yes-or-no computations in the query model, where researchers count requests for information about an input rather than every operation a computer performs. The paper says that, for a quantum algorithm using a given number of queries, a deterministic classical algorithm can approximate its probability of answering yes on all but a chosen small fraction of inputs. For fixed accuracy and a fixed allowance for exceptional inputs, the classical query count grows as a fixed power of the quantum query count.
In practical terms, the result says that for most uniformly random inputs, a classical algorithm can estimate the quantum algorithm’s answer probability without requiring exponentially more input reads. It does not say the classical algorithm can reproduce every quantum computation efficiently, nor does it rule out quantum advantages on problems with useful structure.
PSI said in a company blog post that its internally developed research system used instances of a non-public Anthropic model and found a complete proof during the third day of a run. The paper says its main theorem and its formulation of the conjecture were formalized in Lean. Lean’s kernel checked the formal proof steps; the report also says researchers reviewed the work. These checks support the stated mathematical result, while the account of how the proof was discovered is PSI’s description of its own process.
A Bound on Quantum Input Reads
The result clarifies a boundary between quantum and classical algorithms. It supports the idea that large quantum advantages on yes-or-no problems are not generally available across unstructured random inputs in this query framework. That gives researchers a sharper theoretical reason to focus on problems with mathematical or physical patterns that quantum algorithms can exploit.
It does not settle whether any particular quantum machine will outperform conventional computers on a commercially useful task. The theorem counts input queries, not the full cost of computation, hardware requirements, error correction, or the time needed to process data. Its importance is therefore theoretical: it narrows one route to exponential speedups without deciding the practical value of quantum computing as a whole.
Products Worth Considering
Kindle Paperwhite Signature Edition 32GB (2024) – Auto-adjusting front light, wireless charging, weeks of battery life – Metallic Black
Bose Lifestyle Ultra Wireless Home Speaker
DJI Mini 3 Drone with DJI RC Controller
As an affiliate, we earn on qualifying purchases.
From the 2008 Conjecture to Lean
The conjecture is named for Scott Aaronson and Andris Ambainis, and the paper traces it to a 2008 lecture by Aaronson. It links a question about quantum algorithms to a mathematical statement about polynomials. Earlier work showed that a quantum algorithm’s probability of answering yes can be represented by a formula whose complexity is constrained by how many times the algorithm queries its input. The new proof builds on that connection and earlier results in mathematics and theoretical computer science.
The query model provides a controlled comparison: a classical query reads a selected input bit, while a quantum query can act on a superposition of positions. This does not mean a quantum computer simply reads every position at once; the algorithm must use the resulting information to produce an answer. Known quantum approaches can exploit structure. Shor’s factoring algorithm, introduced in 1994, uses the period of a repeating function. The conjecture addresses whether comparable exponential advantages can arise when the input offers no such favorable pattern.
“The research system found a complete proof during the third day of a run.”
— Physical Superintelligence, in its accompanying blog post
Practical Speedups Remain Open
The theorem does not determine which real-world applications will benefit from quantum computers, or whether those benefits will survive the costs of building and operating practical machines. It also does not bound total computation time: a limit on input reads alone cannot establish that a classical algorithm will finish quickly overall.
The supplied report describes the paper as posted on GitHub and says researchers reviewed it, but it provides no independent review details or publication in a peer-reviewed journal. PSI’s description of the AI system’s contribution is a company account; the source material does not specify how much of the proof was generated by the model, what human researchers contributed, or whether other research groups have independently checked the result.
Independent Review and Follow-Up
The next steps are for other researchers to examine the paper’s argument, assess the Lean formalization, and test whether the theorem’s bounds can be improved or extended. Further work may sharpen the mathematical limits or study related settings beyond the specific query framework reported here.
For quantum algorithm research, the result points attention toward problems with exploitable structure, but it does not identify a winning application. The available source material does not give a schedule for journal publication, independent verification, or a follow-up announcement from PSI.
Key Questions
What does the Aaronson-Ambainis conjecture address?
It asks whether quantum algorithms can achieve exponential advantages on most randomly generated inputs for yes-or-no problems, in a framework that counts how often an algorithm queries its input.
Does the proof show classical computers can match all quantum computers?
No. The reported theorem concerns approximating a quantum algorithm’s probability of answering yes on most inputs in the query model. It does not establish that classical computers can efficiently reproduce every quantum computation.
What did Lean check?
The paper says the main theorem was formalized in Lean and checked by Lean’s kernel, which verifies the logical steps encoded in the formal proof.
How much of the proof came from AI?
PSI says its research system used instances of a non-public Anthropic model and found a complete proof during the third day of a run. The source material does not specify the model’s exact contribution or the human researchers’ role in developing the proof.
Does the result rule out useful quantum computers?
No. It limits a measure of input reads for a defined class of problems, not total computation time or the commercial value of quantum machines. Quantum advantages may still be possible on problems with exploitable structure.
Source: thequantuminsider.com
