In 2024, AI certainly has been one of the most discussed hot topic across the industries. There is a growing trend of overestimating the capabilities of Artificial Intelligence (AI) and its potential applications. This high expectations sometimes create an illusion of a AI as a perfect tool or an individual, making it a disconnect between the realms of possibilities. In this newsletter, we will touch on some of the key insights about the AI and discuss the questions you might already have so stay tuned for what is ahead.

A race for productivity

Since the release of chatGPT, there has been a paramount of expectations that AI will be making large contributions to productivity increase. The fear of missing out for streamlining automation is quite strong. Everyday we can see the news of companies across industries entering to the AI arms race, to secure the top position in their businesses. Some of them have already made into the spotlight for slashing out a large chunk of their workforce and replacing them with AI automation for productivity. The focus question is: Is AI the solution to our challenges?

At Varon Consulting, emerging AI technologies are strategically applied to boost our productivity within relevant contexts. For example, we use it for copywriting and early stages of ideation to get a comprehensive overview of the projects that we are working on. The idea of productivity increased from AI is a polarizing concept that most people certainly have thought about when they first learn about AI autonomy. The other side of it is if someone offered a service that is solely relied on the AI outputs, what are the value proposition for the client and who will be held accountable when AI make an error in the crucial projects? Therefore it need a balancing act of human judgement alongside the large language models’ suggestions to effectively harness the power of the technology.

Here’s an outlook of the AI and operational productivity from the Reuters.

Blackrock CEO Larry Fink claims, AI will “transform margins across sectors”.

Goldman predicts AI will boost productivity growth by up to 3 percentage points per year in the US over the next decade.

Mckinsey insight also predicts that –

Generative AI could add the equivalent of $2.6 trillion to $4.4 trillion annually across the 63 use cases.

Current AI use cases such as automation of repetitive tasks are a positive boost to the productivity. AI can give data-driven insights and also provides a personalized experience. We can definitely see the productivity increase happening in the future. Today’s AI has gotten much more capable since the earlier released versions, however it still falls short of performing tasks at the highest level that many people expect. AI is a helpful addition as a powerful tool in the professional toolbox. However, it won’t be replacing people out of their jobs.

Bottlenecks are also present in the current stage of AI delivery infrastructure.

A.I. could make up 0.5 percent of worldwide electricity use by 2027, or roughly what Argentina uses in a year.

The quoted New York Times article highlights the resources required to produce and maintain AI model services are not very sustainable the AI infrastructure rollout. Similarly, Goldman Sachs report shows that the machinery and energy cost could potentially outweigh the productivity gains. The concerns on the AI is that it is consuming too much electricity for a slight boost in productivity as it also tend to hallucinate. From our research, we reason that AI is not a silver bullet to solve every problems but it can be a leverage to increase productivity margins with properly trained subject matter experts.

The Artificial General Intelligence (AGI), is a vastly different stage of technology today’s AI unlike the tools we know like chatGPT and midjourney. The AGI will have its own cognitive reasoning, learning and problem-solving capabilities like an own entity. When the AGI arrives, then it will make sense to partially automate the white collar jobs for further increase in productivity. AI models at current stage are just performing prediction based text synthesis and it won’t be today or tomorrow yet. The development of Artificial General Intelligence (AGI) is a complex and challenging task, and many experts believe that it will take significant time. Even with AGI, people will need to apply intellectual cognitive reasoning on the AI decision outputs because a machine’s version of best actions will not necessarily equate to the human’s best interests.

If an AI arms race make companies invest massive capital just to maintain the market share, smaller players will inevitably be squeezed out. Industries will shift towards oligopoly and competition will be suffer. If it were to follow the earlier technology investment trends such as blockchain and web3, then productivity will slump even more. Therefore, knowing the capabilities of the technology and its limitation is important to make an operation for the business. The potential benefits are there for those who are able to capitalize and capture the market with innovation. Therefore, we will be seeing companies making capital investments on the AI.

AI learning: Can it teach humanity?

Technology-assisted learning is evolving rapidly, blurring the lines between human teachers and machine algorithms. AI-assisted tools like Writable, Photomath, Google’s Socratic, and Khan Academy’s Khanmigo offer specialized support in various subjects ranging from creative writing to math. You might wonder how an AI ultimately “teach humanity” to humans.

The situation is a bit complex. AI-powered tools could potentially increase the learning outcome of the students. The bright side about learning from the AI is that students can learn without without the fear of judgment. A common challenge that learners face in the education system is being judged from their learning process. Students can challenge the views from AI outputs or have the AI explain to them about a topic a couple times without being judged. By eliminating the fear of judgement from the peers and instructors, AI can improve the learning interactions to support their education.

Another area AI learning can improve on student learning is the critical thinking and AI literacy. This department can be a byproduct of current AI development levels where response inaccuracies are quite common. By challenging the views and adapting to the AI application, students will be well prepared for the critical thinking skills as well as learning the subject matter of the class.

Leaving teaching solely to the machines is not an ideal solution, which is a fact we all can agree. The challenge lies in finding the optimal blend of human and AI guidance like in many other AI use cases. Critics surrounding AI in education often revolve around potential misuse. For instance, a student could rely on bots and take shortcuts to get answers instead of mastering the subject.

A bigger challenge to the AI for education is the AI’s capacity to truly motivate and engage students. AI can provide information and guidance. However, its ability to ignite passion and sustained interest in challenging subjects remains questionable. Essentially, we have witnessed that school children often require hands-on support from a teacher. A classroom devoid of an adult human would not bring the same learning experience to them as the youngsters usually learn behavioral and psychological lessons from the adults.

Did a calculator replace the role of human teachers in math classrooms?

Knowing the limitations and having the education on AI literacy is another important measure to take. AI’s tendency to “hallucinate,” generating incorrect information from its training data in here, could misguide their learning and behaviors. Addressing these issues requires human supervision and teachers are the personnel . Adapting to the technology advances such as the calculator did and teaching how to apply the tools properly would be a key theme in the education for coming years.

Referenced discussions from MIT and edweek/UTA

How AI is streamlining the operations

Artificial intelligence is rapidly reshaping the landscape of digital marketing, empowering businesses to connect with customers in unprecedented ways. By harnessing AI’s predictive power and data-driven insights, marketers are creating a new era of personalized experiences, streamlined workflows, and enhanced customer satisfaction.

Machine learning algorithms can analyze through large amounts of customer data, to uncover consumer’s patterns and preferences. This method of the AI empowers companies to craft highly targeted advertising campaigns that align with individual consumer’s preferences. Take Netflix and Amazon as leading examples for offering tailored experiences through data analysis. Their AI-powered recommendations keep users engaged by suggesting tailored content and products uniquely appealing to the customers.

AI-powered content creation tools are also leading a new era of content creation and automation. These sophisticated AI models can generate compelling text content, from creating blog posts and social media captions to website copywriting. This ensures a consistent flow of fresh, engaging content and free up the valuable time for the employee to focus on high impact projects

Many news organizations like Washington Post and Wall Street Journal, and Politico and Business Insider relies on the AI to help with news article writing, enabling journalists to prioritize in-depth reporting. New York Times, which sued against OpenAI for the damages before, is developing an AI tool to generate headlines for articles with the technology from the latter.

In customer assistants roles where a human assistant can only reply one at a time and might require downtime, AI-powered chatbots and virtual assistants can provide 24/7 support. They can answer frequently asked questions and resolve basic issues instantly. This not only streamlines customer service operations but also enhances satisfaction by delivering prompt and personalized assistance.

As the AI technology continues to evolve, its impact on digital marketing will become greater in terms of value creation. It will help promote more sophisticated ways of connecting with customers and pioneer data-driven strategies for meaningful impacts.

Here’s a helpful example prompt to try on:

“Create a tone of voice guide for COMPANY. To do that follow this procedure:

Enlist tonal cues that apply to different marketing channels
Write best practices for tone of voice examples
End each section with examples of tone of voice in action.”

prompt source: Copyai