Generative artificial intelligence (GAI) is increasingly integrated into financial decision support, yet evidence on its practical role in stock investing remains fragmented. This study develops and evaluates a Deepseek-assisted investment workflow that combines news interpretation, earnings-call summarization, risk factor extraction, and analyst-style narrative generation with conventional quantitative portfolio rules. The case study is designed around A-share and U.S. large-cap equities in a rolling monthly setting, where Deepseek-generated textual signals are transformed into structured factors and integrated with momentum, value, and volatility controls. Results from the study indicate that GAI con-tributes most at the signal construction and research-efficiency layers: it improves event understanding speed, enhances the timeliness of qualitative signals, and supports scenario-based risk diagnostics. How-ever, model hallucination, prompt sensitivity, and data leakage risks can distort investment outcomes if governance is weak. The paper proposes a human-AI collaboration framework emphasizing traceable prompts, cross-source verification, and model risk controls. Overall, Deepseek can be a high-value copilot for stock investment research when it is embedded in auditable and risk-aware processes rather than treated as an autonomous trading agent.

Xi Jin. Applying GAI in the Stock Market Investments: A Case Study of Deepseek. Academic Journal of Business & Management (2026), Vol. 8, Issue 4: 9-16. https://doi.org/10.25236/AJBM.2026.080402.

[1] Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need. Advances in Neural Information Processing Systems, 2017, 30.

[2] Radford A, Wu J, Child R, et al. Language Models are Unsupervised Multitask Learners. OpenAI Technical Report, 2019.

[3] Brown T B, Mann B, Ryder N, et al. Language Models are Few-Shot Learners. Advances in Neural Information Processing Systems, 2020, 33: 1877-1901.

[4] Ouyang L, Wu J, Jiang X, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 2022, 35: 27730-27744.

[5] Araci D. FinBERT: Financial Sentiment Analysis with Pre-trained Language Models. arXiv:1908. 10063, 2019.

[6] Yang X, Zhang Y, Wang H, et al. FinBERT: A Pretrained Language Model for Financial Communications. arXiv preprint arXiv:2006.08097, 2020.

[7] Malo P, Sinha A, Korhonen P, et al. Good Debt or Bad Debt: Detecting Semantic Orientations in Economic Texts. Journal of the Association for Information Science and Technology, 2014, 65(4): 782-796.

[8] Gu S, Kelly B, Xiu D. Empirical Asset Pricing via Machine Learning. Review of Financial Studies, 2020, 33(5): 2223-2273.

[9] Tetlock P C, Saar-Tsechansky M, Macskassy S. More Than Words: Quantifying Language to Measure Firms’ Fundamentals. Journal of Finance, 2008, 63(3): 1437-1467.

[10] Kelly B, Pruitt S, Su Y. Characteristics Are Covariances: A Unified Model of Risk and Return. Journal of Financial Economics, 2019, 134(3): 501-524.

[11] Financial Stability Board. Artificial intelligence and machine learning in financial services: Market developments and financial stability implications. FSB Report, 2017.

[12] Bank for International Settlements. Generative artificial intelligence and machine learning in central banking. BIS Papers No. 149, 2024.