TL;DR
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A paper by David Turturean of Physical Superintelligence reports a proof of the Aaronson-Ambainis conjecture, an open question dating to 2008 about quantum advantages on random inputs. The proof was formalized in Lean and checked by its kernel, but the result concerns input queries, not total computing time or the commercial value of quantum computers.
A paper by David Turturean of Physical Superintelligence (PSI) reports a proof of the Aaronson-Ambainis conjecture, a question open since 2008 about whether quantum algorithms can gain exponential advantages on most random inputs. The result, formalized in Lean and checked by its proof kernel, establishes a bound on input reads in a particular computational model; it does not show that classical computers can match quantum computers in total running time.
The conjecture links a mathematical question about functions to the prospects for quantum speedups. In the query model, algorithms are compared by how many times they access the input rather than by counting every operation involved in a computation. The paper says that for yes-or-no computations, a classical algorithm can approximate a quantum algorithm’s probability of answering yes on all but a small, chosen fraction of uniformly random inputs, without requiring an exponential increase in queries.
According to the report, the paper’s main theorem and its formulation of the conjecture were formalized in Lean. Lean’s kernel checks the logical steps encoded in a formal proof. PSI also said an internally developed research system, using instances of a non-public Anthropic model, found a complete proof during the third day of a run. That account describes the system’s role; the mathematical result is presented in the paper and subjected to formal checking and researcher review.
The bound concerns how often an algorithm reads data, not the full cost of computing an answer. It therefore does not establish that classical computers can simulate quantum computers efficiently in general, nor does it determine the practical value of quantum machines. The paper’s stated result is limited to the query framework and to most inputs under the specified random-input setting.
A Bound on Random-Input Advantages
The result gives researchers a sharper theoretical statement about where to look for dramatic quantum speedups. If a quantum algorithm’s advantage depends on accessing an unstructured, uniformly random input, the theorem says that, in the query model, a classical algorithm can approximate its yes-or-no behavior on most such inputs with a query count that grows as a fixed power of the quantum query count, for fixed accuracy and a fixed allowance for exceptional inputs.
That finding supports a distinction central to quantum algorithm research: substantial advantages may depend on structure that an algorithm can exploit, rather than arising automatically from quantum processing. It gives theorists a boundary to work against when analyzing candidate algorithms. It does not identify which real-world tasks will yield useful speedups, or show that all quantum advantages require the same kind of structure.
The practical stakes remain indirect. A query bound does not include all computation, hardware constraints, error correction, or the time and resources required to prepare and process data. The theorem strengthens a theoretical framework, but does not settle the commercial case for quantum computing.
Products Worth Considering
Sonos Era 100 SL Compact Microphone-Free Speaker, White
Amazon Kindle Scribe (64GB) with Premium Pen, Metallic Jade
DJI Mic 3 Wireless Microphone Kit (2 Transmitters, 1 Receiver, Charging Case)
As an affiliate, we earn on qualifying purchases.
From 2008 Conjecture to Formal Proof
The conjecture is named for Scott Aaronson and Andris Ambainis and is traced in the paper to a 2008 lecture by Aaronson. It arose from work connecting quantum algorithms to mathematical statements about polynomials. Earlier research established a link between a quantum algorithm’s probability of answering yes and a polynomial whose complexity is limited by the number of input queries. The new proof addresses the conjecture built around that connection.
The query model simplifies a computer’s interaction with data so researchers can compare information access independently of other computational costs. A classical query reads a bit at a selected position; a quantum query can act on a superposition of positions, though it does not simply reveal every bit at once. The model is useful for proving limits, but its scope is narrower than a comparison of complete machines.
Known quantum algorithms also illustrate why structure matters. Shor’s factoring algorithm, introduced in 1994, uses a repeating mathematical pattern to find a period. The conjecture asks what can be said when favorable patterns are absent and inputs are randomly generated. The reported proof addresses that question in the query setting rather than resolving every question about quantum computing.
“An internally developed research system, using instances of a non-public Anthropic model, found a complete proof during the third day of a run.”
— Physical Superintelligence, in its accompanying blog post
Limits of the Query-Model Result
The result’s precise implications depend on the assumptions and bounds in the paper, including the chosen accuracy and the permitted fraction of exceptional inputs. It addresses query complexity on most uniformly random inputs; it does not provide a general bound on total computing time or cover every input distribution and computational task.
The supplied report says the proof was reviewed by researchers, but does not identify the reviewers or give details of any independent review process. It also does not establish how the AI system’s contribution was divided among generating ideas, constructing arguments, and producing the final formalization. Those details matter for evaluating the research process, while the formal Lean check addresses whether the encoded proof follows from its formal steps.
It remains unclear which applications, if any, will be affected directly by the theorem, and whether additional results will tighten the mathematical bounds. The finding does not establish that useful quantum speedups are absent; it leaves open advantages tied to structure and other measures of computational cost.
Further Proof and Application Work
The immediate next step is scrutiny of the paper’s statement, assumptions, and formalization by researchers in quantum complexity and formal verification. Readers can compare the paper’s mathematical argument with the Lean-checked version, while further independent review can help clarify the proof’s scope and any qualifications.
On the research side, theorists can investigate whether stronger bounds are possible and how the theorem applies to specific algorithmic settings. Work on practical quantum computing will still need to show that a task has exploitable structure and that a quantum implementation offers an advantage after costs beyond input queries are counted. No application-level speedup is established by the result alone.
Key Questions
What does the proof establish?
It establishes the Aaronson-Ambainis conjecture in a query-complexity framework: for a quantum algorithm, a classical algorithm can approximate its yes-or-no acceptance probability on most uniformly random inputs without an exponential increase in input queries, under fixed accuracy and exceptional-input allowances.
Does this mean quantum computers have no speed advantage?
No. The result limits input-query requirements in a specific setting. It does not rule out practical quantum advantages, particularly on problems with exploitable structure, and it does not compare total computing time.
What role did AI play?
PSI said its internally developed research system, using instances of a non-public Anthropic model, found the proof during the third day of a run. The paper reports that the theorem was formalized in Lean and checked by Lean’s kernel. The precise division of work between the system and researchers is not detailed in the supplied report.
What is the query model?
It compares algorithms by the number of times they access an input, rather than by counting all computational operations. That makes it useful for studying information access, but it does not measure the full runtime or practical cost of a computation.
What happens next?
Researchers can examine the paper and its formalization, test the result’s assumptions, and pursue stronger bounds or applications to specific algorithms. The proof itself does not establish a commercial quantum-computing use case.
Source: thequantuminsider.com
