Technological Netscape Moments: An overview of recent years and how we got to this point.

Technology is the major driver of progress in the world—think: fire, the wheel, electricity, engines, the telephone, cars, etc. and recently: computers, the internet, artificial intelligence. 

Let’s look at netscape moments in the last 70 odd years. 

The term ‘Netscape Moment’ is used in this article to denote the transition point where a technology becomes essential to the general public and when a complex technology is made simple enough for the average person to use (e.g. browsing the web instead of using command-line tools)

1956: Twenty two years before home computers, the term "artificial intelligence" was coined by American computer scientist John McCarthy. It was proposed during the Dartmouth Summer Research Project on Artificial Intelligence—the birth of AI as an academic field.

1978: Personal computers began to enter homes and workplaces.

1993: The general public became aware of the emergence of the World Wide Web.

1995: The real netscape moment: the introduction of the browser. The beast was already there; one could now start to browse it. (Foreshadowing the advent of AI—ChatGPT was just the browser.) This moment served as the "Big Bang" of the internet era, signaling to investors and the public that the World Wide Web was becoming a central, highly valuable, and mainstream technology.

2000: Deep learning –> machine learning –> neural networks. Note: deep learning uses networks with many "hidden" layers (often dozens or hundreds) between the input and output layers. Each layer feeds its output into the next, allowing for a "layered architecture" where information is gradually distilled, enabling the model to learn abstract, hierarchical representations of data.

2015: One billion dollars was pledged to develop artificial intelligence for ‘the benefit of humanity.’ 

2016: Advent of transformers; a type of neural network that learns from huge data sets resulting in full text responses, understanding the underlying meaning of text and starting off the era of the algorithm.

2016 - 2023: Known as the pre-training era which resulted in Pandora’s Box being opened when AI models were taught to generate code and given access to the internet.  

2023 - 2025: Post-training era.

2025 -        : Recursive Self-Improvement era. Models rewrite their own code; the parent literally rewrites the child. By 2026, it was reported that some AI models were capable of writing 100% of the code for specific tasks. 

The public now has access to frontier Artificial Intelligence models. (Claude 4.6, ChatGPT 5, Gemini 3 Pro.)

So what do we have at this moment in time (beginning 2026) after a mere 20 years of intense development? 

We have: trained, autonomous computer programs that use algorithms to recognise patterns in large datasets.They have agency capabilities to predict, make decisions and generate content, without human intervention, using tools like LLMs. 

This is called the biggest information surge in history. They also have access to sets of rules and protocols that allow different software programs to communicate with each other and share data and functionalities. They are able to learn, talk to, answer back, adapt and collaborate with other agent networks on the internet.  

Now ‘netscape moments’ appear almost on a daily basis, change is happening exponentially, the curve is turning vertical— automation is following the same kind of curve as physical law; slow then fast, then all at once. 

Here is a question to ponder on going forward:

Will free public access to AI models continue? If it is a possibility that AI models will soon become strategic assets and might, therefore, be kept internal, won’t it be wise to use AI now to build and to increase oneself as an asset, while it is still publicly accessible?