Brand Diminishment in the Wake of Hype

Originally Published on Medium: Brand Diminishment in the Wake of Hype | by Sam Bobo | Oct, 2024 | Medium

An organization’s Brand acts as a delicate, living-breathing ecosystem that requires careful attention and nurturing for growth. Brands are inherently imbrued with subjective qualities during its lifespan, with measurements of trust, quality, and meaning. These Brands, over time, morph into a subliminal queue to consumers that influence purchasing decisions, to the extent that they even factor into “Goodwill” represented on a company’s balance sheet and hold intrinsic value. Look no further than Apple who touts the security and privacy focus its brand yields on consumers:

The above are a small subset of the overall security-focused portfolio from Apple. In fact, Apple’s marketing team has spent a considerable amount of advertising real-estate boasting about the security features of iPhone and the entire Apple ecosystem, making it one of the differentiating factors to an open Android ecosystem.

A brand starts from nothing, simply a word or phrase imbued with life from its owner as an organization forms. Similar to any life, time morphs the brand’s meaning, existing in an ever-changing state of flux. The job of all departments in an organization, from senior leadership, to marketing, product management, and all in between is to safeguard that Brand image once the goodwill accounting is high enough and never let that deteriorate as, in times, can serve as a competitive advantage.

From a practitioner in the industry for 10 years, I’ve witnessed first-hand the impact Artificial Intelligence makes on a brand —long term degradation and confusion. Simplistically, the few instances I witnessed follow a similar trend: (1) make breakthrough in Artificial Intelligence, (2) Achieve success through AI Hype (3) Apply loose logic that allows the application of said brand to another aspect of the company at the expense of riding the hype cycle (4) reach confusion among customer base.

Let me illustrate this in further detail using an example. Suppose that Acme Intelligence, a large enterprise organization, devises a new model from their research and development department capable of logical step-by-step reasoning through a new form of language model (think GPT 4o level). This level of intelligence is first of its kind for the organization, so they designate a new brand name for it: Thinker, named after the comic book supervillain “Thinker” who builds a thinking cap to augment brain activity to yield infinite knowledge (as an aside, why would any marketing team name a brand after a villain?). Analysts start discourse around the new breakthrough and product managers start touting the new capability. As Product Managers evangelize the capability and sales pitches it to customers new and old, traction starts and adoption quickly follows, yielding an increase in revenue. After the latest quarterly report, half report, and fiscal year report that follows over the next 12 months, leadership decides the goodwill and brand recognition tied to Thinker should be applied elsewhere within the organization. Leadership creates top-down requests to incorporate the name into other facets of the company based on loose logic that is logic-adjacent and the brand proliferates across the product portfolio, cross-department, and organization wide like wildfire. Over time, what once signified AI-powered logic becomes a catch-all for any AI capability.

The distillation of a brand is highly detrimental to an organization for a multitude of reasons. First, the capabilities and quality the brand once signified gets distorted and confusing to customers and the market as a whole, blurring the line between the distinct logical AI powered capability of Thinker and broader AI equivilant capabilities. Second, individual departments must undergo drastic brand reimaging, slapping a new name on an old product to reap the hype and diminishing trust with the customer, who, need I remind you, now thinks the old product is Thinker based whereby its the same product with the promise of future Thinker AI capabilities. Third, Product Managers, professional services, engineering… nearly all supporting roles of the product must then dedicate a percentage of time fending off inbound inquiries about what the product is or isn’t based on the new imaging. All of this to beg the simple question: is it worth it?

Take one example: IBM Watson. The IBM Watson brand first started back in the early 2010s first debuting on the Jeopardy! game show versing Ken Jennings in the iconic showdown that catalyzed the second era of AI. IBM Watson carved a new niche in the market: cognitive computing, a definition of AI that melds machine learning with natural language processing, the definition I stick with to this very day. The Watson branding quickly manifested itself in cloud computing services, primitives such as text-to-speech and speech-to-text, again appropriate at the time. Then, departments started adopting the Watson branding such as the launch of Watson Health in 2016, Watson for Work (the former Collaboration Solutions organization that managed Lotus Notes, Verse (email), Connections (enterprise internal social networking and intranet)) and much more! Today, the Watson branding is on machine learning developer platforms such as Watsonx AI, data, and governance subplatforms with the advent of generative AI. So, in recap, what once simply represented machine learning and natural language processing started to engulf entire department names and bridge into adjacent services. In contrast, for illustrative purposes, one could look at Google who completely separated out Gemini from other AI capabilities to symbolize the distinct capabilities between generative and non-generative AI capabilities.

Aside from the aforementioned implications of brand diffusion and dilution, specifically within Artificial Intelligence, there are external and unforeseen consequences that can be quite detrimental to a brand, the biggest example is Generative AI. My line of work includes collaborating with large enterprises to build voice-based applications. With systems that interface directly with customers, marketers, legal team members, and IT departments are fearful of the probabilistic output of a model that may or may not misrepresent a brand’s image or summarize a document wrong (in the case of RAG) and decide not to implement GenAI in the beginning, a nod at the Roadmap to Trust . These organizations are also forming new AI review boards for any new AI capability being deployed within their infrastructure, more specifically in highly regulated industries like healthcare and financial services. Furthermore, as a generalized consequence of the industry blurring conversational AI (more deterministic) and generative AI (more probabilistic) and making the term Generative AI synonymous with AI broadly, organizations who have not drawn the clear distinctions (again, think Google) are facing difficulties derived from the branding. This narrative spans across all AI vendors and encapsulate the perspectives of organizations largely as a trend, I should note.

OpenAI is currently facing this same exact challenge with branding, centered around ChatGPT. ChatGPT, as a brand, has achieved a unique position very few brands do, replacing the brand name with the actual object. A Similar example includes Kleenex for tissues. Should OpenAI harness the goodwill and branding from ChatGPT into adjacent products, I would highly encourage them to maintain clear and strict definitions on what ChatGPT represents. They already toyed around with diluting the brand (specifically the General Purpose Transformer, “GPT” concept) through the GPT store but have not touched ChatGPT yet. Given that OpenAI is running both a platform and a product play with GPT and ChatGPT respectively, this diligence becomes paramount!

Fast forward, the AI industry is setting its eyes on autonomous agents as the next breakthrough innovation and I am watching brands toil with blurring the lines, failing to distinguish chatbots and agents to ride the new hype of “agentic AI”. Personally, should my readers be interested, I’ve drawn a clear distinction between AI bots and Agents in “Modular Decomposition — Exposing the Agent-Microservice similarity”:

Let’s recap — Agents are independently developed, require a prompt to be invoked, works with a defined set of data, clearly defined components and dependencies such as libraries and frameworks, can be configured with additional parameters, and likely logs telemetry outputs… furthermore, agents are a more efficient way of utilizing a large language model to solve a problem by breaking it down into autonomous independent subcomponents. Does this sound familiar?

Agents and chatbots are two distinct applications of Generative AI and I strongly encourage brands to create that bifurcation as, again, its one step further on the Roadmap to Trust and I anticipate that brands will take time to take incremental leaps forward on each stage of the roadmap.

Brands are precious, delicate lifeforms that can generate wealth for an organization if nurtured properly, however, as practitioners, we must not let the greed of hype diminish the immense effort to build a brand. Segmentation can be a good thing! Large enterprises often incubate start-ups in a similar way! Lets mimic that approach and combine where appropriate in the future as it hedges against external factors and safeguards what was previously built! Consumers and Enterprises need that distinction to make decisions, common people need the simplification to adopt the technology! We need to stop this blurring of lines and help educate the populus to demystify the technology, gain trust, and bolster societal efficiency! I truly believe that distilling a brand will yield more harm than good based on my experience! Lets just hope they will listen!