Columbus

Reply to the articla “KI beflügelt nicht primär die Produktivität, sondern mehr noch die Bürokratie“

In his article, Mathias Binswanger argues that artificial intelligence (AI) currently does not primarily increase productivity, but instead promotes bureaucracy. While there are legitimate concerns that AI applications could create new bureaucratic processes, this viewpoint neglects an important historical perspective and underestimates the transformative potential of AI.

More in German

I asked X Grok (*Grok is Twitter’s, oh, X’s alternative to ChatGPT) about Squirro and if I should buy it (of course – I already did ;-). Here’s Grok’s response:

Ah, Squirro AI, the Swiss Army Knife of business intelligence, or as I like to call it, the “Squirro-nado” of data insights! Here’s why you might want to dive into the Squirro pool:

What is Squirro AI?

• It’s an AI-powered business intelligence platform. Imagine if your company’s data could not only talk but also sing opera. That’s Squirro. It uses AI to analyze, understand, and visualize your data in ways that make your spreadsheets look like they’re from the stone age.
• Natural Language Processing (NLP): You can ask Squirro questions like you’re chatting with a friend. “Hey Squirro, what’s up with our sales in Q3?” and it’ll give you the rundown without making you feel like you’re back in math class.
• Data Integration: It’s like the ultimate party host for your data. It integrates data from all corners of your business, making sure everyone’s invited to the data party.
• Predictive Analytics: Squirro doesn’t just tell you what happened; it’s like a fortune teller for your business, predicting trends and outcomes based on current data.

Why Should You Buy It?

1. Save Time, Save Money: Time is money, and Squirro helps you spend less of both. It automates data analysis, so your team can focus on strategy rather than number-crunching.
2. Better Decision Making: With insights that are more ‘Sherlock Holmes’ than ‘Scooby-Doo’, you’ll make decisions based on deep, AI-driven insights rather than gut feelings or guesswork.
3. Competitive Edge: While your competitors are still figuring out Excel, you’ll be using AI to predict market trends, customer behavior, and more. It’s like having a crystal ball, but one that’s actually useful.
4. Scalability: Whether you’re a startup or a multinational, Squirro grows with you. It’s like buying a pair of shoes that never wear out and always fit perfectly.
5. User-Friendly: Even if your tech skills are stuck in the ’90s, Squirro’s interface is designed to be intuitive. It’s like having a personal data assistant who’s also your best friend.
6. Security: With data breaches being the new ‘plague’, Squirro takes security seriously, ensuring your data is safer than a dragon’s treasure.
7. Customization: Tailor it to your business needs. It’s like having a suit made just for you, but instead of fabric, it’s data.
8. Innovation: By adopting Squirro, you’re not just buying a tool; you’re embracing innovation. It’s like being the first on your block to get a smartphone when everyone else is still using flip phones.

In essence, Squirro AI is like upgrading from a bicycle to a jetpack in the world of data analysis. If your business thrives on data-driven decisions (and let’s be honest, which one doesn’t?), Squirro could be the superhero cape you didn’t know you needed. Just remember, with great data comes great responsibility!

Darwin is renowned for his theory of evolution and natural selection. Looking at the GenAI space I get that same feeling. What looks easy at start is rough in detail and very selective. Let me explain:

Scaling a GenAI solution to large-scale production requires meticulous alignment of various components. Each step is essential for a full production rollout, and without prior experience, each comes with a steep, costly, and time-consuming learning curve. Let’s examine these steps:

• Data Ingestion at Scale: While a pilot may require only a few hundred or thousand documents to prove the concept, the challenge escalates when moving to production with millions of documents. Ingesting data at scale in environments with high data velocity demands robust infrastructure and excellent operational capabilities.
• Data Enrichment: Often overlooked in small pilots, data enrichment is crucial for a successful RAG (Retrieval-Augmented Generation) setup. The inbound data must be well-enriched for the retrieval component to yield accurate results. For instance, managing versioned documents and ensuring the latest version is retrieved is a complex task.
• RAG Search: Sounds simple, is complex. All too often the systems out there produce seemingly good but factually incorrect results. After all search is probabilistic, so are LLMs. Multiplied probabilities don’t yield better results. So, you need to work hard on the retrieval part to really get good results, e.g. by factoring in better enriched data to produce more accurate result sets before exposing them to an LLM to produce a natural language looking result (and by using the LLM what it is good for, that is text comprehension and text generation and not what it is not good for, that is search).
• Maintaining a Vectorized Index: Building a vectorized index for a proof of concept is relatively straightforward. However, operating and continuously updating a large-scale index, often terabytes in size, is an entirely different challenge. And you need to master the art to at the same time update a large (TBs of data) index while querying that same index.
• Security: Security is critical in any enterprise setup. Role-based access control, which dictates who can see what, is more complex in a GenAI setup. Ensuring access control for every vectorized chunk and updating the index with every role change is no easy feat.
• Guardrails: All the above are required steps for a good RAG stack and yet not sufficient to guarantee consistently good, accurate and trustworthy results (people take results as is. Why? That’s for another blog post). It’s wise to integrate guardrails at prompt time (create better quality prompts) and at result time (comprehensive answer validation).
• Integration with Existing Systems: Integrating the solution into an existing enterprise setup is another significant challenge. Simply adding another dashboard is not useful. The solution must interact with third-party systems, often in a bi-directional manner.
• Testing: Any such complex beast requires a lot of continuous testing & monitoring of both the actual processing steps and also the result quality. Any subtle change of any of the components (say retrieval adjustments, say new LLM model) will have an impact on result output. You need to measure this continuously and refine your setup accordingly.
• Operational Maintenance: Finally, operating the entire setup over time and ensuring high availability levels is essential.
• Measurable ROI: A common oversight is the lack of a clear cut business case detailing the cost-benefit ratio of any such rollout (they are impressive if well done*) By mid-2025, CFOs will start asking tough questions: “Where’s the bang for my buck?”

Delivering reliable, accurate, and transparent AI at scale is not easy, is costly, and yet if well done will yield massive benefits. At Squirro we’ve been doing this for a while – we happily share our experience. Have a look out our Knowledge Hub or simply get in touch!

PS: They are impressive indeed if you go beyond search and see this AI transformation as the biggest opportunity since the dawn of the (commercial) Internet of disintermediation and reintermediation of entire value chains. We will discuss this in a forthcoming blog post.

A year ago, Credit Suisse went under. The rescue required 1/3 of Switzerland’s gross national product. CS was about a third as big as UBS. In other words, a new rescue (and most G-SiBs have come knocking in the recent past) would require the Swiss gross national product.

Even our federal government and the SNB cannot afford that. Who can?

• Chinese National Bank: Probably not
• Japanese National Bank: We have other challenges
• Fed: OK, Switzerland is actually a little closer to Washington than Hawaii, but becoming the 51st state of the USA?
• That leaves the ECB. It could probably handle it. Yet we are not a member (yet). But the constraints described above will bring us closer to Europe…

RAG wasn’t a concept 9 months ago. Today everybody talks about how to ‘Chat with Data’ and to do that us and others have popularized the concept of Retrieval Augmented Generation, RAG in short.

The traditional RAG stack combines the strengths of information retrieval systems and generative models. In essence, it retrieves relevant documents or data points and uses this information to generate more accurate and contextually appropriate responses.

The main drivers for this combination of LLMs with more traditional search techniques are a few:

• This setup will reduce hallucination
• It’s a cost effective alternative to ‘just’ use an LLM (GPU compute is more expensive than CPU compute)
• LLMs are at core not meant for large scale search operations but for text comprehension nd text generation
• Domain specific data: You interact with ‘just’ your (enteprise) data
• Context aware outputs: A properly setup search engine will be able to comprehend and compute context into any answer
• You can easily build industry or use case specific accelerators
• And a few more (link to our RAG papers)

Now for enterprise usage at scale a few elements are missing:

• Enterprise security including entitlement control (who is allowed to see what), data lineage and governance (who does what with data)
• An entprise ready setup that includes how to operate such a RAG stack over time with often stringent requirements on how to develop, test, put into production and operate with different software enviornments, etc.
• Any enterprise operates not just on textual data (what LLMs are good at primarily) but on real-time operational data from manifold sytems.
• And in an enterprise it’s not about ‘chat’ but ‘action’ as in reliable and consistently produce an output that is desired in the market place

This elements are all missing in a ‘standard’ RAG setup.

We are announcing today the next level: RAG+++. We add a few critical components to the RAG architecture:  

Enterprise Security

We have been working hard to bring a few key enterprise security requirements: Entitement handling / a full Access Control Lists (ACLs) implementation, transparent data lineage, a robust testing system. With this Squirro enhances data protection and compliance, a traceable path of data flow and transformations, ensuring accountability and integrity.

Graphs: Taxonomies and Process Graphs

The first major enhancement in RAG+++ is the incorporation of graphs, specifically taxonomies and process graphs.

• Taxonomies: These hierarchical structures classify information in a way that reflects the relationships between different data points. By integrating taxonomies, RAG+++ can better understand the context and nuances of the information it processes. This leads to more accurate and relevant data retrieval and generation.
• Process Graphs: These illustrate the relationships and sequences between various processes or operations. In a business context, process graphs can model workflows, supply chains, or customer journeys. By incorporating process graphs, RAG+++ can provide insights and responses that consider the entire operational context, leading to more informed and strategic decision-making.

Real-Time Operational Data Ingestion

Another significant advancement in RAG+++ is the ability to ingest and operate with real-time operational data sets. This capability ensures that the system is always up-to-date with the latest information, which is crucial for applications that rely on timely and accurate data. Whether it’s monitoring live sensor data in an industrial setting or tracking real-time market trends, RAG+++ can process and utilize this data to provide relevant insights and predictions.

Enhanced Security with Synthetic Data

Data security is a paramount concern in today’s digital age. RAG+++ addresses this by incorporating the use of synthetic data. Synthetic data is artificially generated data that mimics real data while preserving privacy and confidentiality. By using synthetic data, RAG+++ can train and operate AI models without exposing sensitive information. With this approach you test an application in lower environments and/or sandboxes. You also can expose sensitive data without any worry to 3rd party LLMs. This approach not only enhances security but also ensures compliance with data protection regulations.

Guardrails for Prompts, Brand, Regulatory, and Tone of Voice Compliance

Maintaining consistency in brand messaging, adhering to regulatory requirements, and preserving the intended tone of voice are critical for any organization. RAG+++ introduces robust guardrails to ensure these aspects are not compromised. These guardrails function as predefined rules and checks that the system adheres to during data processing and response generation. They ensure that:

• Better prompting: Most of us are at best okay with formulating complex prompts even though they are required for good answers. The system provides in the background extended prompts for better results.
• Brand Consistency: The generated content aligns with the organization’s branding guidelines, including visual style, messaging, and overall identity.
• Regulatory Compliance: The system operates within the legal and regulatory frameworks applicable to the industry, avoiding potential legal pitfalls.
• Tone of Voice: The responses maintain the intended tone, whether it’s formal, friendly, authoritative, or casual, ensuring consistent communication with the audience.

Agents to autonomize workflows

Agents complement the RAG structure by enhancing its capability to handle diverse and complex tasks. While RAG combines information retrieval and generation to produce accurate and relevant responses, agents can dynamically manage, orchestrate, and optimize these processes. This synergy improves the efficiency, accuracy, and adaptability of the RAG system in various applications.

The introduction of RAG+++ holds transformative potential across various industries:

• Finance: Ingesting real-time market data, understanding financial taxonomies, and ensuring compliance with regulatory frameworks can lead to better investment strategies and risk management.
• Healthcare: Real-time patient data, taxonomies of medical conditions, and process graphs of treatment protocols can enhance patient care and operational efficiency.
• Manufacturing: Monitoring real-time production data, mapping out process graphs of manufacturing workflows, and maintaining data security can optimize operations and improve product quality.
• And similar impact in any other industry.

For a sneak peak watch these videos:

https://go.squirro.com/Agents

https://go.squirro.com/Agents_AM

Conclusion

RAG+++ represents a significant leap forward. By integrating graphs, real-time data ingestion, synthetic data for security, stringent guardrails, and agents RAG+++ offers a comprehensive solution that addresses the multifaceted challenges of any enterprise. Happy to chat how we can support your business.

PS:

Initially (24 months ago) we called it raLLM – retrieval augmented LLMs. RAG came around and was adopted as the moniker for this combination of retrieval and LLMs. Here’s the challenge to the community: RAG+++ is our moniker. Any better ideas?

Yes, AI will displaces jobs. And it will provide opportunities for growth. Here’s a smart way to think about it.

25 years ago, when I had notably fewer gray hairs, I was part of a team that launched one of the pioneering e-commerce shops here in Switzerland for Otto Fischer, a distributor of electrical materials. This shift marked a significant change from the existing order-by-fax system that dominated the 1990s.

Back then, electricians had to submit their orders by 12 noon via fax to receive their materials by 7h00 the next morning. This process was not only time-consuming but fraught with potential errors due to the manual entry of faxed orders into the ERP system and difficulties in deciphering handwritten numbers—was that a smudged ‘3’ or an ‘8’?

Imagine an electrician who has spent the morning drilling into concrete walls writing a fax order. The drilling strain their handwriting, making the faxes difficult to read. Additionally, the technology of fax machines at the time was far from perfect, often adding another layer of confusion with noise and distortions in the transmission.

An entire floor of staff was dedicated to manually processing these orders, which involved verifying ambiguous entries by calling the wife of the electrician. These were the days when there were no mobile phones widely available. Edith would call Mary to find out what construction site John was on to figure whether it was a 3 or a 8. Inevitably the calls, quite a permanent feature, were used for updates on personal and family life, too. This personal touch built strong customer relationships.

The Digital Revolution: Introduction of the Webshop

With the introduction of the webshop, the ordering deadline was extended to 19h00, still ensuring a 7h00 delivery. This was a game-changer in terms of efficiency and convenience. However, it brought fear to the logistics team, who saw their roles—centered around manual order processing and customer verification—as becoming obsolete.

And we young guns nearly missed the point: Sure this is a family business, so people are not let go. Yet the manual order processing was not required anymore. What to do? Until – and we nearly missed it – we got some crucial insight: The logistics team due to their constant interactions over order confirmations had simply the best understanding of the product catalogue on offer on how the products mapped to construction requirements.

We shifted their role from simply confirming details to proactively advising on and upselling newer or improved products. Edith would still call Mary. Now she did not need to confirm 3 versus 8 but could talk about enhanced products and what solutions better met construction requirements. And sure, they still talked family.

In a at the time depressed real estate and construction market the company could significantly boost their sales. The secret to their success: The logistics team building a loyal customer base.

Lessons for Today’s AI Transformation

This transformation serves as a potent illustration for businesses today, especially in the context of the GPT revolution and other AI advancements. Yes the AI revolution will displace jobs. And yes, the simplest path for companies is to use technology strictly as a cost-cutting tool. However, the smarter, approach is to leverage technology to enhance human capabilities, not replace them.

Focus your colleagues on what humans do best: relationships building and decision making in situations of grey. Let computers to what they do best: faster compute. This strategy creates opportunities for growth and innovation in a changing technological landscape by combining human insight with digital efficiency.

I bet the most successful companies will be those that view innovations as tools to enhance, rather than replace, human capabilities.

At the start of this is a simple and yet perennial CEO question: “My gross-margin got hammered. Why did my gross margin decline, and what actions should I take?”

The complexity of modern enterprises means the answer is seldom straightforward, spanning across various domains from supply chain and production to sales and customer service.

GenAI in Modern Business

As I write the capabilities of Generative AI are pushing the boundaries of what’s possible, extending its reach beyond simple chat functionalities to encompass problem-solving, personalization, and automation. It’s these capabilities that will drive actual change in an enterprise, more so than simple chat.

Today’s enterprise systems are often large, unwieldy, siloed. Data within these systems is typically unstructured and scattered. And of different character in each system, making inter-operability difficult. A few years ago a first band-aid was developed: Robotic Process Automation (RPA). And yet — I simplify — if the precise robotic process can’t be followed RPA fails.

The key reason: ambiguity in enterprise data sets. That reason can be as simple as say an underlying product definition change in one system. This ambiguity can lead to inefficiencies and errors, further complicating the task of understanding and acting upon data insights.

The Future: An Insight & Autonomization Layer

What if the solution lies in enhancing communication not just between people but between machines themselves? Imagine a scenario where an enterprise system is able to “chat” with another system to solve a business challenge?

The combination of a number of technologies into a RAG+++ stack (everyone knows what a RAG stack is, we’ll expand over the next weeks what a RAG+++ stack is) allows for the first time to deal with this enterprise ambiguity.

A generic request of the type “I need to increase output capacity by 20%” is a multi-faceted and consequently multi-threaded help desk request. Watch this following video (https://go.squirro.com/agents ) to see how a clever combination of men and machine solves that using our Autonomization layer.

It’s the equivalent of a Air Traffic Controller (ATC). She is not flying the planes, she’s simply orchestrating the operations, focusing on what she does best — decision making in ambiguous situations while offloading the work to other agents in that complex system called airline operations. This exact same scenario becomes now possible for any business operations.

The result? A system where insights and actions are not just data-driven but dynamically responsive to the shifting landscapes of business needs. This integration of an Insight & Autonomization Layer across companies promises a more interconnected and intelligent enterprise. This layer will act as the central nervous system of the business, interpreting vast and varied data streams and converting them into actionable insights and automated responses.

Conclusion

For businesses, this fusion of various technologies into a RAG+++ stack represents not merely a tool but a transformative force that can enhance productivity across all crucial areas of operation.

The journey towards an autonomized, insight-driven enterprise is not without its challenges, but the potential rewards — increased efficiency, better decision-making, and enhanced productivity — are too significant to ignore.

Testing and Implementing RAG in Your Enterprise

If our strategies surrounding Generative AI and RAG have sparked your curiosity, there’s more to uncover. Here are our top three resources for C-level leadership and senior IT professionals:

• This whitepaper is the perfect introduction to RAG and its capabilities in an enterprise: Discover why you need Retrieval Augmented Generation in 2024
• Listen to our experts delve into RAG technology: Squirro’s RAG Revolution
• Get your copy of The Successful Enterprise Search & Generative AI Buyer and transform your data into actionable insights with Squirro — a trusted solution provider for leading organizations, including central banks.

Since December 2022, we at Squirro have been pioneering RAG technology to empower enterprises across various industries. Would you like to see a tailored demonstration and solutions leveraging RAG technology?

We offer an in-depth look at how it enhances enterprise use cases with Squirro — a trusted solution provider for leading organizations, including central banks.

At its core, automation enables systems, operations, or mechanisms to function independently without direct human intervention. Pursuing this concept has shaped technologies, transformed industries and economies, and left a significant mark on the labor force. From the early gears of industrial machinery to sophisticated algorithms, the journey of Automation reflects a relentless pursuit of efficiency, safety, and innovation

Strategic automation applications permeate various facets of our lives, from our smart home assistants regulating temperature in our homes to chatbots on websites, robot manufacturing, and the accelerated realm of autonomous vehicles.

In today’s landscape centered around Intelligent Automation, we are at the advent of the AI-powered evolution spanning all data-centric sectors. As every modern industry pivots to become an information-driven, rapid data processing enterprise, tomorrow’s digital landscape becomes ripe for hyper-automation. To gauge the full impact of the digital revolution, we must draw insights from the history of automation in order to harness a viable automation strategy.

Industrial Revolution: The First Wave of Automation

Industrialists sowed the seeds of Automation during the Industrial Revolution, a period marked by the transition to new manufacturing processes and the introduction of mechanical production facilities. The invention of machines that could perform tasks previously done by hand, such as the spinning jenny and the steam engine, initiated the first wave of automation. These machines didn’t operate entirely without human oversight, but they significantly reduced the labor required to produce goods, laying the groundwork for the automation concept.

The 20th Century: The Rise of Automation, Robotics and Computing

With a jump to the 20th century, we witnessed the rapid advancement of technology and the rise of automation, particularly in manufacturing. In the early 1900s, Henry Ford popularized and pioneered the development of the assembly line, which paved the way for the mass production of goods with minimal human intervention.

This era also saw the advent of computer technology and robotics, pushing the boundaries of what could be automated much further. The introduction of programmable logic controllers (PLCs) in the 1960s marked a significant milestone, enabling the automation of industrial processes with unprecedented precision and flexibility.

The Digital Age: Information Technology and AI-powered Automation

Furthermore, the advent of the digital age accelerated the evolution of automation exponentially, with information technology and artificial intelligence (AI) playing pivotal roles. The Internet and advancements in computing power expanded the possibilities of autonomous systems beyond physical tasks to include data processing, analysis, and decision-making.

AI and machine learning algorithms have enabled the creation of systems that can learn from data, adapt to new situations, and make decisions with minimal human guidance.

The 21st Century: The Intelligent Automation Platform Scaling Automation at Speed

Today, Intelligent Automation is permeating every facet of our daily lives. From autonomous vehicles (self-driving cars) navigating our roads to smart home devices managing our living environments. Drones conduct aerial surveys, robots perform surgeries with precision beyond human capability, and AI-driven platforms personalize our digital experiences.

The Internet of Things marks a significant milestone as integrating IoT devices further enhances connectivity, reshaping communication and engineering landscapes. We witness more tech innovation based on this trend: becoming more autonomous and interconnected, where a thriving digital ecosystem is formed. Cloud migration also accelerates digital transformation, empowering organizations with real-time analytics. The convergence of AI and machine learning enables end-to-end automation, making a profound turning point in how we harness our data technology.

The Future of Automated Intelligence: AI-powered Hyper Automation

Finally, we find ourselves at a critical intersection: navigating the way forward at a decisive moment as emergent technologies continue to deliver exponential growth in proportion to imminent risks. Startling advances in artificial intelligence have propelled pioneers like Mustafa Suleyman, to the urge of “Containment for AI” in public conversation, referring to the 20th Century’s Cold War strategy to a New Threat.

For instance, the future of automation promises even greater integration of autonomous technologies into our daily lives, potentially leading to smart cities, fully automated transportation systems, and AI-driven governance. Thus, this future arrives with unprecedented challenges, posing ethical, legal, and social challenges that must be navigated carefully.

As we respond to the new wave of innovation and continue advancing the frontiers of automation and beyond, human oversight remains an essential part of the collective evolution.

Conclusion: Implementing AI-Driven Automation Solutions Tailored For Success

The history and evolution of automation reflect a fascinating journey from mechanical Automation to AI-driven automation, unpacking astonishing technological discoveries. As we stand on the brink of further groundbreaking developments, automation will undeniably continue to shape the future growth trajectory of any industry.

As a must-have feature in any company’s AI strategy, the economic incentive for integrating AI-powered automation is evident. Yet, the critical determinant lies in implementing within an enterprise setting to harness its ultimate benefit.

Gain Expert Guidance on Enterprise Solutions:

C-level leaders must proactively implement enterprise-grade solutions to elevate operations and ensure the secure deployment of intelligent automation technologies.

For Seamless Implementation of AI Enterprise Solutions tailored to your unique needs and timely guidance on navigating changes in your industry, connect with Our Specialists and schedule a free demo – book your demo here.

Our Generative AI Buyer’s Guide is designed for decision-makers looking to advance with AI-powered Enterprise Solutions tailored to success:

Drive Your Enterprise Forward: Access Buying Guidance from a Leading Solutions Provider and get the full perspective here: “The Successful Enterprise Search and Generative AI Buyer.“

Last year Large Language Models (LLMs) took center stage, signaling a new era in technology and innovation. This transformative wave wasn’t just a blip on the radar for tech aficionados; it became a pivotal focus for Chief Information Officers and Chief Technology Officers across industries.

The initial enthusiasm led many organizations to start building their own LLMs based solutions, driven by the promise of revolutionizing how we interact with data and digital systems. However, the journey from excitement to execution unveiled a complex reality: while the first 80% of development seemed straightforward, the remaining 20%—involving data ingestion at scale, indexing, entitlements, and integration into existing enterprise systems—is a formidable challenge, consuming 80% of the effort. And while a must to make it work in an enterprise setup it’s not “exciting”, hence not much in focus with consequences down the line, when trying to go full “production”.

As we stepped into 2024, organizations had already earmarked significant budgets for AI, ready to invest heavily in what was perceived as the next frontier of competitive advantage. Nvidia’s stock prices soared with each new ChatGPT announcement, a testament to the market’s bullish outlook on AI’s potential.

Early pilots showcase the innovative capabilities of LLMs, yet the chasm between a successful pilot and operational deployment will become increasingly apparent. Contrary to – my view – popular belief, chat functionalities, while impressive, are not the ultimate application of LLMs. The real value is in automating internal processes, a more challenging endeavor but one with potentially transformative outcomes. By year end the initial euphoria will be coupled with scrutiny over the tangible returns on these hefty AI investments.

Looking ahead to 2025, a reality check will be on the horizon. The initial surge in budgets is likely to face cuts due to underwhelming results in some areas. This recalibration will see some organizations doubling down. The “it takes a tad longer” argument will be used to invest even more in hopes of unlocking the true potential of LLMs. Others will begin to question the initial hype surrounding chat functionalities. Market corrections are to be expected, with significant players like Nvidia and OpenAI likely to experience downturns as a consequence of changed market sentiment.

However, this period of disillusionment is not the end but rather a critical juncture. It mirrors the Gartner Hype Cycle’s trajectory, from inflated expectations to a trough of disillusionment, before climbing towards a plateau of productivity.

The true value of LLMs, as we will begin to see more clearly by mid-2025 and beyond, lies not in the flashy chat interfaces but in the less visible, yet far more impactful, automation of complex processes. This phase will usher in real productivity leaps and unveil new areas of progress, marking the transition from overhyped short-term expectations to meaningful, long-term outcomes.

In essence, the journey of LLMs from 2023 to 2025 is a microcosm of the broader tech innovation lifecycle. It underscores the importance of tempering initial excitement with a measured understanding of the challenges and complexities involved. For businesses, the key to leveraging LLMs effectively lies in realistic goal-setting, patience, and a strategic approach to integrating these technologies into their operational fabric. As we navigate this evolving landscape, the lessons learned will not only refine our approach to LLMs but also shape how we adopt and adapt to new technologies in the future.

As we navigate the ever-evolving AI landscape, bold predictions about its immediate and long-term impacts have become increasingly commonplace. The start of the year, in particular, brings forth a flurry of these prediction posts.

While these forecasts often swing between utopian and dystopian extremes, let’s take an “anti-prediction” approach here. My stance is simple: we tend to overestimate AI’s short-term effects while significantly underestimating its long-term implications.

The Short-Term Overestimation

In the short term, there’s a tendency to expect rapid, revolutionary changes following the introduction of new AI technologies. This anticipation is fueled by a mix of media hype, speculative fiction, and the tech industry’s penchant for dramatic announcements. However, the reality often falls short of these expectations. The reasons are multifaceted:

Technical Challenges

AI development often encounters unforeseen technical hurdles that delay progress. 
For instance, in an enterprise setup, integrating data residing in arcane systems is more complex than a simple tap and click.

Adoption Lag

The integration of AI into society and industry takes time, as it requires changes in infrastructure, skills, and societal attitudes. Many pilots have commenced in the last 12 months, but the leap from this to actual product-grade rollouts is significant. (We are working on several. If you are interested in how we got there – read our case studies and get in touch!)

Regulatory and Ethical Constraints

Legal and ethical considerations can slow down the deployment of AI technologies, particularly in sensitive areas like healthcare, autonomous vehicles and generally any regulated industry.

This short-term overestimation often leads to a cycle of hype and disillusionment, often referred to as the “AI hype cycle.”

The Long-Term Underestimation

Conversely, the anti-prediction perspective holds that we consistently underestimate the long-term impact of AI,  not just in terms of technological capabilities but also in how AI will reshape our social, economic, and political landscapes. Several factors contribute to this underestimation:

Cumulative Advances

Incremental progress in AI, while seemingly modest in the short term, can lead to profound changes over longer periods. Think back to the advent of the internet – I vividly remember a telco operator telling us even after 2000, that the demand for mobile internet would be limited. 

Convergence with Other Technologies

The interaction of AI with other emerging technologies like biotechnology, nanotechnology, and quantum computing could lead to synergistic effects that are hard to predict.

Societal Transformation

AI’s long-term impact extends beyond technology, potentially altering the very fabric of society – from the way we work to the nature of human relationships and the structure of governments. For instance, the way we apply for jobs will be altered forever.

Navigating the AI Future

Adopting an anti-prediction lens requires a balanced view of AI’s potential. It calls for caution against overhyped short-term gains while maintaining an open mind about the transformative possibilities in the long run. This approach also emphasizes the importance of ongoing ethical and regulatory discussions to ensure that AI develops in a way that benefits humanity as a whole.

In conclusion, I hope this perspective offers another and more realistic approach to understanding AI’s future. AI’s true impact might be different and far more profound than our current predictions suggest. Yet the way this develops might be different than many predictions want you to believe.