AI Fin ResearchFinance research is being automated. News, guides and a vendor map on the AI tools, data and techniques changing the field, for researchers and students.https://aifinresearch.com/A new frontier model ships every 10 days, and no university is built to keep uphttps://aifinresearch.com/news/2026-10-09-a-new-frontier-model-every-10-days/https://aifinresearch.com/news/2026-10-09-a-new-frontier-model-every-10-days/Anthropic, OpenAI and Google released at least 28 language models between 1 January and 9 October 2026. Higher education staff name the pace of change as their top AI challenge.Fri, 09 Oct 2026 00:00:00 GMTAACSB's 2026 standards never mention AI. That is a gigantic mistake.https://aifinresearch.com/news/2026-10-09-aacsb-2026-standards-do-not-name-ai/https://aifinresearch.com/news/2026-10-09-aacsb-2026-standards-do-not-name-ai/Its own research says 97 percent of schools already cite AI in accreditation materials and nearly half lack the governance to manage it. The standards are silent.Fri, 09 Oct 2026 00:00:00 GMTThe standalone academic data platform is running out of reasons to existhttps://aifinresearch.com/news/2026-10-09-academic-data-platform-query-layer/https://aifinresearch.com/news/2026-10-09-academic-data-platform-query-layer/AI assistants now ship with vendor data inside and recognize the licenses a user already holds. The query screen is the part of the platform that does not survive.Fri, 09 Oct 2026 00:00:00 GMTAcademic research is far behind the AI frontier, from finance to mathematicshttps://aifinresearch.com/news/2026-10-09-academic-research-behind-ai-frontier/https://aifinresearch.com/news/2026-10-09-academic-research-behind-ai-frontier/A paper about a model is out of date before it clears review. In mathematics, the newest results come from a model no academic can run.Fri, 09 Oct 2026 00:00:00 GMTAI has arrived for finance research, and data vendors cannot police ithttps://aifinresearch.com/news/2026-10-09-ai-has-arrived-vendors-cannot-police-agents/https://aifinresearch.com/news/2026-10-09-ai-has-arrived-vendors-cannot-police-agents/Agents now operate a computer better than the human baseline on the standard test. Licenses forbid most AI use of data, and almost none of it can be detected.Fri, 09 Oct 2026 00:00:00 GMTAI is obliterating the pipeline from finance programs to industry jobshttps://aifinresearch.com/news/2026-10-09-ai-is-obliterating-the-pipeline-from-finance-programs-to-industry/https://aifinresearch.com/news/2026-10-09-ai-is-obliterating-the-pipeline-from-finance-programs-to-industry/Data vendors and AI companies now sell to banks and asset managers directly. The junior tasks that absorbed finance graduates are the first ones their products automate.Fri, 09 Oct 2026 00:00:00 GMTResearchers in China, Japan and Korea run whole papers through coding agents and post howhttps://aifinresearch.com/news/2026-10-09-asia-researchers-run-papers-through-claude-code/https://aifinresearch.com/news/2026-10-09-asia-researchers-run-papers-through-claude-code/Their own forums and blogs describe pipelines of 30 agents, a seven-week paper with 108 experiment runs, and an eight-hour draft. Every account keeps the checking for a person.Fri, 09 Oct 2026 00:00:00 GMTIf the platform forbids agents, build the dataset yourself from the sourcehttps://aifinresearch.com/news/2026-10-09-build-your-own-dataset-from-the-source/https://aifinresearch.com/news/2026-10-09-build-your-own-dataset-from-the-source/Filings, financial statements and earnings calls were public before any vendor packaged them. An agent can go back to the source, and the tools to do it are free.Fri, 09 Oct 2026 00:00:00 GMTDevelopers are leaving Copilot and editor plugins for Claude Code and Codexhttps://aifinresearch.com/news/2026-10-09-developers-are-leaving-editor-plugins-for-claude-code-and-codex/https://aifinresearch.com/news/2026-10-09-developers-are-leaving-editor-plugins-for-claude-code-and-codex/In six months Claude Code went from 18% to 39% of professional developers and Codex from 3% to 16%, while Copilot and Cursor fell. Price, tools and computer use explain it.Fri, 09 Oct 2026 00:00:00 GMTKeep paying for call transcripts, but only for the history, the identifiers and the rightshttps://aifinresearch.com/news/2026-10-09-pay-for-transcripts-in-the-age-of-agents/https://aifinresearch.com/news/2026-10-09-pay-for-transcripts-in-the-age-of-agents/An agent now does the cleaning, splitting and scoring that used to justify a data budget. What it cannot produce is history, identifiers and the right to use the text.Fri, 09 Oct 2026 00:00:00 GMTResearchers are using AI against the rules, because being left behind is worsehttps://aifinresearch.com/news/2026-10-09-researchers-use-ai-against-the-rules/https://aifinresearch.com/news/2026-10-09-researchers-use-ai-against-the-rules/At a major 2026 conference, more than one in five reviewers who were told not to use an LLM used one anyway. The fear driving that is rational.Fri, 09 Oct 2026 00:00:00 GMTOpenAI releases machine-produced math results with Lean proofs and compute figureshttps://aifinresearch.com/news/2026-10-06-openai-math-results-release/https://aifinresearch.com/news/2026-10-06-openai-math-results-release/The results come from an internal model. OpenAI says the average result used compute equal to roughly three hours of ChatGPT Pro thinking.Tue, 06 Oct 2026 00:00:00 GMTA benchmark of nine factor-mining methods finds no approach consistently winshttps://aifinresearch.com/news/2026-10-03-factorbench-automated-factor-mining/https://aifinresearch.com/news/2026-10-03-factorbench-automated-factor-mining/FactorBench compares roughly five thousand machine-mined factors across five equity markets, from genetic programming to LLM agents.Sat, 03 Oct 2026 00:00:00 GMTSelf-evolving research agents show no consistent gain from accumulated skills in a factor testhttps://aifinresearch.com/news/2026-09-27-evermine-self-evolving-alpha-research/https://aifinresearch.com/news/2026-09-27-evermine-self-evolving-alpha-research/Across 18 long-horizon alpha-research runs and 48 continuation branches, evolved capabilities did not reliably beat the starting set.Sun, 27 Sep 2026 00:00:00 GMTSeven LLMs scoring the same earnings calls agree with a rank correlation of 0.52https://aifinresearch.com/news/2026-09-25-llm-text-measures-diverge-across-models/https://aifinresearch.com/news/2026-09-25-llm-text-measures-diverge-across-models/Across thirteen text measures on S&P 500 transcripts, the choice of model changes the size, sign and significance of downstream coefficients.Fri, 25 Sep 2026 00:00:00 GMTLLM literature reviews hold up for broad claims and break down for paper-level oneshttps://aifinresearch.com/news/2026-09-23-measurement-error-ai-literature-reviews/https://aifinresearch.com/news/2026-09-23-measurement-error-ai-literature-reviews/A test on economics papers that use rainfall as an instrument finds accuracy falls as the reading task needs more context.Wed, 23 Sep 2026 00:00:00 GMTAgents can propose investment factors, but a frozen referee has to judge themhttps://aifinresearch.com/news/2026-09-22-propose-dont-judge-factor-referee/https://aifinresearch.com/news/2026-09-22-propose-dont-judge-factor-referee/A working paper splits factor research in two and finds that a referee the agent cannot touch admits 5 to 11 times fewer false factors.Tue, 22 Sep 2026 00:00:00 GMTScientists report saving nearly 7 hours a week with AI, and the bottleneck moves to verificationhttps://aifinresearch.com/news/2026-09-15-ai-in-science-early-insights/https://aifinresearch.com/news/2026-09-15-ai-in-science-early-insights/A study of 15 million Gemini interactions, over 2,600 specialized models and a survey of over 600 scientists maps how AI is used in research.Tue, 15 Sep 2026 00:00:00 GMTDaiwa Securities moves its Sakana AI consulting platform from trials into production developmenthttps://aifinresearch.com/news/2026-08-05-daiwa-sakana-ai-platform-enters-production/https://aifinresearch.com/news/2026-08-05-daiwa-sakana-ai-platform-enters-production/Sakana AI, the Tokyo lab behind The AI Scientist, is building a total-asset consulting system for Daiwa under a partnership signed in October 2025.Wed, 05 Aug 2026 00:00:00 GMTAACSB's AI framework for business schools grows to 84 schools in its July 2026 updatehttps://aifinresearch.com/news/2026-07-28-aacsb-ai-framework-84-schools/https://aifinresearch.com/news/2026-07-28-aacsb-ai-framework-84-schools/The report series went from 26 institutions in July 2025 to 48 in January 2026 and 84 in July, and it names faculty development as the critical success factor.Tue, 28 Jul 2026 00:00:00 GMTThe one public review of WRDS's Ask AI chatbot finds a lookup tool with a six-question limithttps://aifinresearch.com/news/2026-06-10-wrds-ask-ai-chatbot-first-public-review/https://aifinresearch.com/news/2026-06-10-wrds-ask-ai-chatbot-first-public-review/Librarians in Singapore tested the beta and found it useful for locating variables, unreliable on first try and capped at six questions a session.Wed, 10 Jun 2026 00:00:00 GMTKorea's financial regulator loosens network separation so firms can use AIhttps://aifinresearch.com/news/2026-05-25-korea-loosens-network-separation-for-ai/https://aifinresearch.com/news/2026-05-25-korea-loosens-network-separation-for-ai/The Financial Services Commission will exempt firms with strong security from the rule that keeps their internal systems off the internet, and it is reviewing full repeal.Mon, 25 May 2026 00:00:00 GMTStock Claude Code and Codex wrote 117 papers unaided, and none cleared a top venue's barhttps://aifinresearch.com/news/2026-05-18-claude-code-codex-wrote-117-papers-none-accepted/https://aifinresearch.com/news/2026-05-18-claude-code-codex-wrote-117-papers-none-accepted/With only a light scaffold, Claude Code matched the average human ICLR 2025 submission on manuscript review. Scores fell once reviewers opened the code behind each paper.Mon, 18 May 2026 00:00:00 GMTA paper-writing pipeline for Claude Code went from 6,400 GitHub stars to 51,000 in five monthshttps://aifinresearch.com/news/2026-05-17-taiwan-built-paper-pipeline-for-claude-code-51000-stars/https://aifinresearch.com/news/2026-05-17-taiwan-built-paper-pipeline-for-claude-code-51000-stars/academic-research-skills runs 32 agents across ten stages with two integrity gates. Its builder says the checks caught 15 fabricated citations in one real paper.Sun, 17 May 2026 00:00:00 GMTMirae Asset moved research assistant work to AI and made the assistants junior analystshttps://aifinresearch.com/news/2026-05-13-mirae-asset-moves-research-assistant-work-to-ai/https://aifinresearch.com/news/2026-05-13-mirae-asset-moves-research-assistant-work-to-ai/The Korean broker renamed its research unit the AI Research Center and expects an analyst to cover 20 to 30 stocks, up from 10 to 15, BizWatch reports.Wed, 13 May 2026 00:00:00 GMTChinese brokers put DeepSeek-V4 into use within hours of its releasehttps://aifinresearch.com/news/2026-04-27-chinese-brokers-deploy-deepseek-v4-within-a-day/https://aifinresearch.com/news/2026-04-27-chinese-brokers-deploy-deepseek-v4-within-a-day/Guotai Haitong deployed the model on release day, 24 April 2026, and Industrial Securities reports full integration in two hours, according to National Business Daily.Mon, 27 Apr 2026 00:00:00 GMTIn 912 AI-written economics papers, the ideas trail human work far more than the executionhttps://aifinresearch.com/news/2026-04-03-ai-economics-papers-ideas-trail-execution/https://aifinresearch.com/news/2026-04-03-ai-economics-papers-ideas-trail-execution/Against 41 papers from the AER and AEJ: Economic Policy, a study puts 71% of the quality gap on the research idea and finds no significant gap on robustness.Fri, 03 Apr 2026 00:00:00 GMTChina's quant funds are turning their trading compute into AI labshttps://aifinresearch.com/news/2026-03-23-chinese-quant-funds-build-ai-labs/https://aifinresearch.com/news/2026-03-23-chinese-quant-funds-build-ai-labs/Shanghai Securities News names four quant managers with their own institutes, supercomputing centers or open source models, and asks which will produce the next DeepSeek.Mon, 23 Mar 2026 00:00:00 GMTReview of Finance bans hidden prompts in manuscripts and sets a one-year penaltyhttps://aifinresearch.com/news/2026-03-23-review-of-finance-llm-policy/https://aifinresearch.com/news/2026-03-23-review-of-finance-llm-policy/The journal's LLM policy also requires authors to disclose any AI use in the cover letter and in the manuscript, and limits referees to supportive use.Mon, 23 Mar 2026 00:00:00 GMTJapan's financial regulator tells firms not to hold back on AI for fear of riskhttps://aifinresearch.com/news/2026-03-03-japan-fsa-tells-firms-not-to-hold-back-on-ai/https://aifinresearch.com/news/2026-03-03-japan-fsa-tells-firms-not-to-hold-back-on-ai/The FSA's revised AI paper says customer-facing use, almost absent in late 2024, is now being launched in limited form, and that firms need not solve every problem first.Tue, 03 Mar 2026 00:00:00 GMT