{"id":9059,"date":"2026-10-06T16:37:33","date_gmt":"2026-10-06T11:07:33","guid":{"rendered":"https:\/\/www.tisatech.in\/blog\/?p=9059"},"modified":"2026-10-06T16:37:34","modified_gmt":"2026-10-06T11:07:34","slug":"how-does-generative-ai-work","status":"publish","type":"post","link":"https:\/\/www.tisatech.in\/blog\/how-does-generative-ai-work\/","title":{"rendered":"How Does Generative AI Work? A Step-by-Step Guide"},"content":{"rendered":"\n

Open ChatGPT, Gemini, or Meta AI on WhatsApp, type one line, and an answer appears in seconds. Ask for a poster, and you get a picture. Ask for a code fix, and you get working code. Students, freelancers, and shop owners across India now do this every day. But how does generative AI work behind the screen?<\/p>\n\n\n\n

The short answer:<\/strong> Generative AI does not understand information exactly the way people do. During training, it processes large amounts of text, images, code, audio, or other data to learn patterns. When you send a prompt<\/a>, the system converts it into numerical representations, processes those representations through a trained model, and generates an output step by step.<\/p>\n\n\n\n

In this guide, you will follow each step with simple examples, from tokens and transformers to the reason AI sometimes gets facts wrong.<\/p>\n\n\n\n

\n \n \n
\n\"WhatsApp\"\n\n<\/div>\n \n \n
\n Book a Free Demo Class\n<\/div>\n \n \n\n Chat Now\n<\/a>\n \n<\/div>\n\n\n\n

Why is everyone talking about generative AI right now?<\/h2>\n\n\n\n

Since ChatGPT launched in late 2022, generative AI has moved from tech news into daily life. One reason is how easy these tools are to use. People no longer need technical skills to get useful results from AI. Students use AI tools to plan assignments, freelancers use them to draft proposals, and shop owners use them to design posters in minutes. The wide range of everyday uses has made generative AI relevant to people far beyond the tech industry.<\/p>\n\n\n\n

In a Google\/Ipsos survey from late 2025, 83% of people in India said they had used generative AI in the past year. Only the UAE and Nigeria scored higher among the 21 countries surveyed. Rutgers University’s GenAI Adoption Tracker<\/a> follows working-age Indians and shows the same trend. Its authors say the figures are upper-bound estimates, because online surveys reach people who already use AI.<\/p>\n\n\n\n

What is the main idea behind generative AI?<\/h2>\n\n\n\n

Generative AI learns patterns from huge amounts of data. Then it uses those patterns to create something new when you give it a prompt. It does not simply retrieve a ready-made answer from its training data. Instead, it uses learned patterns and the information in your prompt to generate an output. Microsoft’s AI 101 guide<\/a> explains the same idea in a beginner-friendly way.<\/p>\n\n\n\n

You can understand the process in three broad steps:<\/p>\n\n\n\n

    \n
  1. Training:<\/strong> The model processes large amounts of text, images, audio, or other data and adjusts its parameters to capture useful patterns.<\/li>\n\n\n\n
  2. Representation:<\/strong> The data is converted into numbers that the model can process and relate to one another.<\/li>\n\n\n\n
  3. Generation:<\/strong> When you enter a prompt, the trained model uses what it learned to predict and generate new content step by step.<\/li>\n<\/ol>\n\n\n\n

    What data does generative AI learn from?<\/h2>\n\n\n\n

    Different models learn from different types of data. Text models learn from books, articles, websites, and code. Many of them use Common Crawl<\/a>, a free public archive of web pages. A 2024 Mozilla Foundation study found that at least 64% of 47 large language models released between 2019 and 2023 trained on filtered versions of it. Image models learn from pictures paired with short captions. The LAION-5B dataset<\/a>, for example, holds 5.85 billion image-text pairs. Audio and video models learn from speech, music, and clips that show how things move.<\/p>\n\n\n\n

    More useful data can help a model learn a wider range of patterns, but data quality matters just as much. Duplicates, low-quality pages, harmful content, and other noise can affect what a model learns. Raw web data is messy, so teams clean it before using it for training. They remove junk, duplicates, and other low-quality content. One Google-led study<\/a> found that a single 61-word sentence was repeated more than 60,000 times in a popular training dataset. Examples like this show why cleaning and filtering training data matters.<\/p>\n\n\n\n

    <\/svg>Also Read<\/div>
    What is Generative AI and Why Students Should Learn It?\n \n\n<\/div><\/div>
    <\/svg><\/div><\/div><\/a>

    What are tokens and why do they matter?<\/h2>\n\n\n\n

    Tokenization is the process of breaking text into small pieces called tokens.<\/strong> The system assigns each token a number from a fixed vocabulary. A token can be a whole word or just part of one. Common words like \u201cthe\u201d usually stay whole, while rarer or longer words may split into smaller parts, so \u201clearning\u201d may become \u201clearn\u201d and \u201cing.\u201d This subword approach<\/a> keeps the vocabulary manageable and lets the model handle words it has never seen before.<\/p>\n\n\n\n

    Next, the model represents each token as a longer list of numbers called an embedding. These numerical representations help the model capture relationships between tokens and their context. Tokens used in similar contexts often have more similar representations. “Happy” and “joyful” sit close together, while “happy” and “table” sit far apart.<\/p>\n\n\n\n

    A model has a token limit that applies to the input and output combined. Many tools also charge by the token. One English token is often around four characters or about three-quarters of a word, although the exact number varies by tokenizer and text. Hindi and other non-English texts usually need more tokens per word, so the same message takes up more room. You can paste your own text into OpenAI’s tokenizer tool<\/a> and see exactly how it splits.<\/p>\n\n\n\n

    Images follow a similar idea. The system cuts a picture into small patches and turns each patch into numbers, much like words.<\/p>\n\n\n\n

    How does generative AI learn from data?<\/h2>\n\n\n\n

    Learning happens through a simple guessing game. The model sees a sequence of tokens with the next token hidden and tries to predict it. Then it compares its prediction with the actual next token. If the prediction is wrong, it measures the error and adjusts tiny numbers inside itself, called parameters or weights, to do better next time. This adjustment process uses a method called gradient descent<\/a>. The model keeps adjusting its parameters in a direction that reduces its error.<\/p>\n\n\n\n

    The model repeats this process across large amounts of data, gradually improving its predictions. Meta trained its 405-billion-parameter Llama 3.1 model on over 15 trillion tokens, using more than 16,000 GPUs<\/a>.<\/p>\n\n\n\n

    What do neural networks and transformers do?<\/h2>\n\n\n\n

    A neural network is a stack of connected layers. Each layer has many small units that take numbers in and pass numbers out.<\/p>\n\n\n\n

    Older networks, called recurrent networks, read text one word at a time. That made them slow to train and weak on long sentences. In 2017, Google researchers published Attention Is All You Need<\/a>, which introduced the transformer. It dropped the word-by-word reading and made training much faster. Most modern AI tools now use this design.<\/p>\n\n\n\n

    Attention lets a transformer decide which words in a sentence matter most for understanding each word. In “The bank near the river was calm,” the model focuses on “river” to know that “bank” means a riverside, not a financial institution.<\/p>\n\n\n\n

    Transformers use attention across tokens to capture relationships in complex inputs. Different attention heads can focus on different relationships.<\/p>\n\n\n\n

    <\/svg>Also Read<\/div>
    Generative AI vs Traditional AI: What Should Students Learn in 2026?\n \n\n<\/div><\/div>
    <\/svg><\/div><\/div><\/a>

    How does generative AI predict the next word?<\/h2>\n\n\n\n

    When you type a prompt, the system turns it into tokens, then into numbers. Those numbers pass through many transformer layers. At the end, the model scores every possible next token, and the scores become probabilities. Take the prompt “The sun rises in the.” The model might assign a high probability to \u201ceast\u201d and a much lower probability to unrelated words such as \u201ckitchen.\u201d As 3Blue1Brown explains<\/a>, the model assigns a probability to every possible next word instead of picking one with certainty.<\/p>\n\n\n\n

    The model doesn’t always pick the most likely next word. Always choosing it (greedy search) works for short outputs but gets repetitive in longer ones<\/a>. So the model uses sampling, picking randomly but weighted by probability. A setting called temperature controls how adventurous this is: low gives predictable replies, high gives more variety. That’s why the same prompt can give a different reply each time.<\/p>\n\n\n\n

    How does generative AI generate content from a prompt?<\/h2>\n\n\n\n

    Let’s follow one prompt from start to finish. Say you type: “Explain photosynthesis to a class 9 student.”<\/p>\n\n\n\n

      \n
    1. Tokenize:<\/strong> The system splits your sentence into tokens and gives each one a number.<\/li>\n\n\n\n
    2. Embed:<\/strong> It turns each token into an embedding, so the model can read the meaning.<\/li>\n\n\n\n
    3. Process:<\/strong> The transformer layers use attention to link the words. The model sees that “class 9 student” means it should keep the answer simple.<\/li>\n\n\n\n
    4. Predict:<\/strong> The model scores every possible next token and picks one by sampling. For this prompt, it may pick “Photosynthesis” first.<\/li>\n\n\n\n
    5. Repeat:<\/strong> It adds that token to the text and predicts the next one. This loop continues until the answer is complete.<\/li>\n<\/ol>\n\n\n\n

      A specific prompt gives the model more context and steers its predictions toward a more focused output, and the Claude prompting guide<\/a> makes the same point about being clear and direct.<\/p>\n\n\n\n

      Compare these two prompts:<\/p>\n\n\n\n