AI academic search engine Consensus has completed a $11.5 million Series A funding round led by Union Square Ventures, aiming to accelerate academic search efficiency. It has a database of over 200 million papers covering all fields of science and academia, supporting various search methods, including keywords, conceptual associations, and Q&A. It uses custom language models to re-rank search results, providing the top 20 most relevant entries. It integrates features like citation generation and supports integration with tools like Zotero, catering to students and researchers.

Ant Group, in collaboration with Xin京Bao and Beike Finance, has developed and launched a generative large-scale model product named "Beike Finance · Smart Guidance for ESG." This product aims to provide users with more accurate, comprehensive, and valuable analysis and interpretation of ESG (Environmental, Social, and Corporate Governance) reports of A-share listed companies. It can answer various user queries in real-time and offer in-depth analysis and summaries."Beike Finance · Smart Guida

According to Citigroup, artificial intelligence has the potential to replace more jobs in the banking sector, more so than in other industries. This technology is expected to disrupt consumer finance and enhance workers' productivity. Based on a new report on artificial intelligence released by the bank on Wednesday, about 54% of jobs in the banking industry have a high potential for automation, while an additional 12% of positions could be augmented by this technology.

Two AI anchors from CCTV Finance's new media have started working during the Two Sessions, able to answer questions in the fields of economy, industry, and tourism 24/7. The AI anchors are supported by the Ant Group Digital Human Platform, featuring high modeling accuracy, complete dynamic details, natural pronunciation, and high fidelity. The Ant Group Digital Human Platform provided technical support for the digital torchbearers relay activity during last year's Hangzhou Asian Games. In 2021, CCTV introduced an AI sign language anchor to provide year-round service for the hearing impaired. Zhu Guangquan has already 'interviewed' the AI sign language anchor with challenging jokes.

Tonghuashun Ask Finance has launched the HithinkGPT large model, featuring a decoder-only architecture based on transformers. It offers five versions: 7B, 13B, 30B, 70B, and 130B, with a maximum allowable text input of 32k tokens. It supports API calls, web embedding, co-construction, and private deployment capabilities. The model demonstrates outstanding performance in the financial domain, passing multiple financial examinations with an average score of 75.9.

Zhaolian Consumer Finance has launched 'Zhaolian Zhilu', the first open source model with 13 billion parameters in the consumer finance industry, which has received authoritative recognition in the field of general knowledge. This model features high precision and financial expertise, meeting customization needs to improve work efficiency and reduce costs. The model will be applied in subdivided scenarios such as risk control management, code generation, and operational enhancement, while continuing to explore intelligent agent technology and promoting technology for good. Zhaolian Consumer Finance's open source model 'Zhaolian Zhilu' has ranked on authoritative lists C-Eval and CMMLU.

📚 Large language models (LLMs) like ChatGPT and GPT-4 face challenges in passing the CFA exam. 📊 Research shows that GPT-4 significantly outperforms ChatGPT but still encounters difficulties, especially in higher-level exams. 🔍 Specialized training and knowledge acquisition targeted at the CFA exam may be key to improving AI's performance in the finance sector.

The Bank of England believes that AI could exacerbate financial stability risks and undermine trust in banks. They are concerned that algorithms may lead to discriminatory decisions and replicate historical biases. The Bank stated that the use of opaque AI models by multiple companies could impact market stability. They recommend that central banks consider how to limit the risks of AI used by financial firms. The Bank's viewpoint has heightened concerns among countries about the rapid development of AI.

Hangseng Electronics has announced that the financial large model product LightGPT has begun internal testing. LightGPT is a large language model designed specifically for the finance sector. Hangseng's financial industry large model features specialization, compliance, and lightweight characteristics. Photon, the intelligent assistant based on LightGPT, provides various application scenarios for financial institutions. LightGPT supports private/cloud deployment and complies with regulatory requirements of the Chinese financial market.

The heat of large models in the financial industry is unprecedented, with banks, insurance companies, and brokerage firms all making moves. However, three major obstacles—immature technology, compliance restrictions, and a lack of use cases—constrain their application. Currently, large models are limited to shallow areas such as intelligent customer service in financial scenarios, remaining far from core business functions. The technology of large models itself has defects in accuracy and judgment, which do not meet the 100% accuracy requirements of the financial sector. The training and use of large models must comply with data security regulations and are affected by compliance restrictions. The promotion of large models requires rich use case support, but currently, financial scenarios are not diverse enough.

The London Stock Exchange Group (LSEG) collaborates with Microsoft to develop a custom AI large model for the financial sector. LSEG aims to leverage generative AI to enhance financial data analysis comprehensively. The application space for generative AI in finance includes business scenarios such as investment strategy recommendations, risk assessment, and stock forecasting.

93% of companies plan to increase their investment in AI over the next 12 months. All Chinese companies and 98% of Indian companies were found to be preparing to increase their generative AI investments. The entertainment, technology, and consumer goods sectors are leading, followed by construction and real estate, financial services, and transportation sectors.