Role of AI In Data Analysis, Rethinking Data Teams



In December’s newsletter I will cover three topics that I see being brought up in recent conversations with the decision makers (CEOs, CIOs, CFOs, COOs) at top the investment management industry here in New York City.

The 3 topics of conversation are,

  1. With all the hype around AI, what does data analytics look like at a top-tier institutional investment firm?
  2. With the talks of recession looming, what is the role of data teams within a institutional investment firm? Should we look into cutting roles and position?
  3. A new research paper from Google confirms what I have been saying about Artificial Intelligence (AI), and Artificial General Intelligence (AGI)…

I hope you enjoy…and thanks for reading!

With All the Hype Around AI, I Believe That Data Analysis Will Be a Distinctly Human Endeavor for a Foreseeable Future

While AI like ChatGPT can generate basic queries and charts, realizing data’s full potential requires human analysts.

Data analysis is more art than science, demanding intuition to make sense of messy, inconsistent datasets and understand their real-world meaning.

No AI today replicates analysts’ versatility in judgment, causal reasoning and teasing nuances from ambiguous information.

Before finding answers, analysts must unravel gnarly data knots.

So while AI will automate rote tasks, it cannot replace analysts’ creative problem-solving and communication abilities. The best systems will augment human skills with AI’s speed to extract exponentially more value from data.

With technologists and analysts partnering, we can combine machine number-crunching with human insight and strategy.

The future is not AI replacing analysts, but augmenting them.

Our ability to find meaning in noise is something machines cannot yet replicate.

Data analysis will remain a distinctly human endeavor.

Rethinking Data Teams as Profit Drivers

I am seeing a shift in the top-down thinking of leading the investment management industry.

They are realizing that with the right optimization, their data teams can evolve from a cost center into a growth driver.

Breaking from Outdated Mindsets

Historically, the investment management industry have viewed their data teams mainly as an expense item – a team of technical specialists to support standard reporting but not directly driving profits or revenue.

That traditional mindset severely under-leverages the tremendous value data can offer in today’s fiercely competitive, fintech-driven industry.

Financial data is enormously powerful (when processed correctly) to enable smarter decisions and customer intelligence. Data expertise thus needs to be elevated into a central business optimization capability.

Optimizing Data Teams to Drive Growth

Top-tier institutions are investing in specialized data engineers, data scientists and visualization experts – and integrating these cross-functional teams directly into critical business units.

By empowering data teams to extract deep insights around risk, lending, personalization and beyond, financial organizations can transform themselves into lean, insight-driven profit engines.

Equally important is providing the data engineering capabilities so that data teams have the technology infrastructure for modeling, experimentation and analysis at speed and scale.

Final Takeaway – Treat Data as a Key Asset

To successfully leverage data as a growth driver, the investment management industry must drop outdated notions of data teams as an IT expense.

Data and data talent must be regarded as a renewable resource and a core business optimization asset.

With that shift in mindset, financial organizations can unlock tremendous competitive advantage even in the face of fintech disruption.

The Recent Research Paper From Google On AI & AGI

In a recent research paper released by Google, critical insights into the world of artificial intelligence (AI) and artificial general intelligence (AGI) have been brought to the forefront.

The findings not only align with existing discussions on AI but also underscore the challenges hindering the realization of AGI.

The study focused on transformer models, such as the ones propelling ChatGPT, revealing a notable limitation in their ability to generalize beyond their training data.

While these models exhibit exceptional performance in tasks directly linked to their training, they encounter difficulties when faced with unfamiliar scenarios and abstract thinking – characteristics inherent in human cognition.

The implications of these limitations are significant.

The research suggests that, despite the fervor surrounding AGI, we are still a considerable distance away from achieving it.

According to Pedro Domingos, an AI expert, transformers, though powerful, are also opaque, leading to potential overestimations of their capabilities.

While they can achieve superhuman proficiency in specific tasks, they lack the adaptability and abstract reasoning capabilities of the human mind.

It’s a call for realistic expectations.

The paper serves as a reminder that today’s AI, while undeniably valuable, has constraints.

The quest for truly general artificial intelligence remains a challenging and elusive goal.

Transformers, while excelling in specific natural language tasks, fall short when it comes to matching the expansive scope of human intelligence.

In light of these findings, the message is clear: we must approach AI development with realistic expectations to foster responsible innovation.

Overhyping prototype technologies can lead to misconceptions about AI capabilities.

However, despite the challenges highlighted, the pursuit of more capable and trustworthy AI continues.

The journey toward AGI may be complex, but the insights gained pave the way for a more informed and measured approach to AI development.

Link to the original research paper –

If you found this valuable, please sign up for my newsletter below.

Signup to my LinkedIn newsletter. No email address needed.

You’ll get exclusive content on digitizing finance, along with best practices for leveraging data and technology to become a more strategic, analytics-driven institutional investment firm.