AINL#017 Augmented Intelligence in Investment Management Newsletter
Welcome to the 017 Edition of the Newsletter on Augmented Intelligence in Investment Management (AINL). Every two weeks, we deliver five unique insights tailored to empower investment decision-makers. Our insights are carefully curated by a seasoned team of market specialists. Unbiased, actionable and practical. They will help you navigate through the noise.
AINL#17 SYNTHESIS
What do these recent developments mean for investment decision-makers?
1. Human-AI Collaboration Drives Client Trust, Not Just Efficiency
Despite the growing technical competence of AI agents (Article 2 | Xu et al., 2025), their current limitations in executing complex, long-horizon tasks autonomously suggest that augmented intelligence—not full automation—remains the superior model for most financial service contexts. Both Le et al. (Article 3 |2025) and Yang et al. (Article 4 | 2025) show that visible human-AI collaboration increases trust, adoption, and client satisfaction, particularly under uncertainty. For investment firms, this implies that embedding AI into client-facing workflows should highlight human oversight rather than aim for substitution, as doing so enhances perceived reliability and accelerates ROI on AI investments.
2. AI Amplifies Risk Vectors, Operationally and Systemically
The FRI study on biosecurity risks (Article 1 | 2025) serves as a warning: frontier LLMs are already exceeding expert expectations in sensitive domains such as virology. This underscores the urgent need for AI-specific operational risk frameworks, especially for asset managers exposed to ESG-sensitive mandates or reputational risk. Meanwhile, Kuhn & Schuller (Article 5 | 2025) highlight how AI-enabled decentralisation is reshaping the financial infrastructure landscape—where governance, not geography, will define future systemic resilience. Investors should prepare for a convergence of technology risk and regulatory arbitrage, particularly in digital assets and algorithmic capital allocation.
3. Alpha Will Depend on Strategic AI Adoption
The takeaway from Fund Forum and WAIB 2025 is clear: AI, Web3, and digital identity will define the next competitive frontier. Yet, Xu et al. (2025) show that even top-tier AI agents can autonomously complete only 30% of professional tasks. This means that alpha in the coming decade will emerge not from AI capabilities alone, but from firms’ ability to integrate these tools within trusted, human-led systems. Building out agent-to-agent infrastructure, anchored in self-sovereign identity and transparent ledgers, may become a foundational strategy for firms seeking a durable edge in both public and private markets.
TOP 5 ARTICLES
ARTICLE ONE
Biosecurity Risks Created By LLMs
ARTIFICIAL INTELLIGENCE | Forecasting Research Institute | 7_2025 | Report
Important Development
FRI releases new study on risks of bio-engineered epidemics. As AI capabilities improve, concerns have grown about the potential biosecurity risks posed by frontier large language models (LLMs). This study systematically assesses expert beliefs about these risks through surveys of 46 domain experts in biosecurity and biology, along with 22 expert forecasters (superforecasters).
Why Relevant to You?
Subject-matter specialists expect risks to rise substantially if LLMs passed key virology-lab competency tests. They also thought LLMs would not attain these capabilities before 2030, but LLMs were already achieving them in 2025. The relevance for investors is self-explanatory.
ARTICLE TWO
Benchmarking LLM Agents on Consequential Real World Tasks
ARTIFICIAL INTELLIGENCE | Frank F. Xu et al. (Carnegie Mellon University, Duke University) | 05 2025 | Article
Important Findings
Interesting research design. How performant are AI agents at accelerating or even autonomously performing work-related tasks? To measure the progress on performing real-world professional tasks, the authors introduced TheAgentCompany, an extensible benchmark for evaluating AI agents that interact with the world in similar ways to those of a digital worker: by browsing the Web, writing code, running programs, and communicating with other coworkers. Result: the most competitive agent can complete only 30% of tasks autonomously.
Why Relevant to You?
This paints a nuanced picture on task automation with LM agents–in a setting simulating a real workplace, a good portion of simpler tasks could be solved autonomously, but more difficult, long-horizon tasks are still beyond the reach of current systems.
ARTICLE THREE
The Future of Work > Understanding the Effectiveness of Collaboration Between Human And Digital Employees in Service
HUMAN & ARTIFICIAL INTELLIGENCE | Le, K. B. Q., Sajtos, L., Kunz, W. H., & Fernandez, K. V. | 01 2025 | Article
Important Findings
The article explores how visible collaboration between human employees (HE) and digital employees (DE), such as AI-powered chatbots, positively influences customer satisfaction. Across five empirical studies, the authors demonstrate that cues signalling coordination, shared goals, and supervisory dynamics enhance perceptions of process fluency and team cohesion. These visible HE-DE interactions boost customer trust and willingness to adopt services, primarily due to the transparency they create around service execution.
Why Relevant to You?
For corporate leaders and financial professionals, this research is highly relevant. It redefines AI not merely as a tool but as a collaborative partner. Transparency and the perception of human-AI teamwork directly affect customer confidence and adoption rates. Investing in systems that showcase collaboration and not substitution can drive stronger client engagement, greater operational efficiency, and improved ROI on AI deployments.
ARTICLE FOUR
How Human Presence Amplifies the Impact of AI Investment Advice
HUMAN & ARTIFICIAL INTELLIGENCE | Yang, Bauer, Li, and Hinz | June 2025 | Paper
Important Findings
This paper marks an important development in understanding how AI can be effectively integrated into financial services. While much of the existing literature focuses on the technical performance of AI systems, this study shifts attention to how clients perceive and act on AI-generated advice in real-world settings. By demonstrating that human involvement enhances not the quality, but the acceptance of AI advice—especially under uncertainty—it uncovers a crucial psychological and behavioural dimension often overlooked in AI deployment.
Why Relevant to You?
The benefits of adding humans to an AI-driven service did not stem from improved advice quality, as one might assume. Instead, they arose from the human element, which enhanced customers’ sense of trust and emotional reassurance. This suggests that the mere presence of a human advisor can act as a powerful signal, increasing the persuasive impact of AI-generated advice.
ARTICLE FIVE
Fund Forum and WAIB 2025 Reflections
HUMAN & ARTIFICIAL INTELLIGENCE | Michiel Kuhn, Markus Schuller, Panthera Solutions | July 2025 | Article
Important Findings
At the end of June 2025, the Principality of #Monaco played host to two notable conference highlights. The IMpower FundForum 2025 brought together the established global investment community, while the WAIB Summit 2025 showcased the emerging voices shaping the future of Web3 and AI. Michiel Kuhn and Markus Schuller are reflecting on their takeaways. In short: A brave new tech-world in search for purpose, while triggering massive legacy issues.
Why Relevant to You?
The flood into private markets has turned niche diversification into systemic vulnerability, as ever-larger pools of capital erode alpha and expose fragile liquidity under stress. At the same time, tariffs, stablecoin crackdowns and the digital-euro sprint aren’t isolated skirmishes but frontlines in a new sovereignty war over code-based money, where fractured rails mean access, not borders, will define finance. And as autonomous AI agents gain real-world agency, trust will hinge on self-sovereign identities and immutable ledgers to govern a burgeoning agent-to-agent economy.
