Artificial General Intelligence, or AGI: A concept suggesting a more advanced version of AI than we know today, one that can perform tasks much better than humans while also teaching and advancing its own capabilities.
AI ethics: Principles aimed at preventing AI from harming humans, achieved through means like determining how AI systems should collect data or deal with bias.
AI safety: An interdisciplinary field that is concerned with the long-term impacts of AI and how it could progress suddenly to a super intelligence which could be hostile to humans.
Algorithm: A series of instructions that allows a computer programme to learn and analyse data in a particular way, such as recognising patterns, to then learn from it and accomplish tasks on its own. For instance, the algorithms used by the new chatbots leverage pre-trained language models to learn from input text and generate human-like responses.
Alignment: Tweaking an AI to better produce the desired outcome. This can refer to anything from moderating content to maintaining positive interactions towards humans.
Anthropomorphism: When humans tend to give nonhuman objects humanlike characteristics. In AI, this can include believing a chatbot is more humanlike and aware than it actually is, like believing it’s happy, sad or even sentient altogether.
Artificial Intelligence, or AI: The use of technology to simulate human intelligence, either in computer programmes or robotics. AI is the star of the show, often portrayed in science fiction as a world-ending technology with killer robots. However, the reality is less dramatic. AI refers to computer technology that aims to make machines work intelligently, similar to the human mind. It involves training computers to learn, respond, and make decisions, although it still relies on learning from humans to reach a high level of performance. While AI is currently a hot topic, the concept has been around for decades. The latest chatbots are a form of AI that can analyse vast amounts of data and generate responses, albeit with some guidance or prompts.
Bias: In regards to large language models, errors resulting from the training data. This can result in falsely attributing specifi characteristics to certain races or groups based on stereotypes.
Chatbot: A programme that communicates with humans through text simulating human language.
ChatGPT: An AI chatbot developed by OpenAI that uses large language model technology.
Cognitive computing: Another term for Artificial Intelligence.
Data augmentation: Remixing existing data or adding a more diverse set of data to train an AI.
Deep Learning: A method of AI, and a subfield of machine learning, that uses multiple parameters to recognise complex patterns in pictures, sound and text. The process is inspired by the human brain and uses artificial neural networks to create patterns. Deep Learning algorithms enable AI to continuously evolve and improve its performance over time.
Diffusion: A method of machine learning that takes an existing piece of data, like a photo, and adds random noise. Diffusion models train their networks to re-engineer or recover that photo.
Emergent behavior: When an AI model exhibits unintended abilities.
End-to-end learning, or E2E: A Deep Learning process in which a model is instructed to perform a task from start to finish. It’s not trained to accomplish a task sequentially but instead learns from the inputs and solves it all at once.
Ethical considerations: An awareness of the ethical implications of AI and issues related to privacy, data usage, fairness, misuse and other safety issues.
Foom: Also known as fast takeoff or hard takeoff. The concept that if someone builds an AGI, it might already be too late to save humanity.
Generative adversarial networks, or GANs: A generative AI model composed of two neural networks to generate new data: a generator and a discriminator. The generator creates new content, and the discriminator checks to see if it’s authentic.
Generative AI: A content-generating technology that uses AI to create text, video, computer code or images. The AI is fed large amounts of training data, finds patterns to generate its own novel responses, which can sometimes be similar to the source material.
Google Bard: An AI chatbot by Google that functions similarly to ChatGPT but pulls information from the current web, whereas ChatGPT currently is limited to data until 2021 and isn’t connected to the Internet.
Guardrails: Policies and restrictions placed on AI models to ensure data is handled responsibly and that the model doesn’t create disturbing content.
Hallucination: An incorrect response from AI. Can include generative AI producing answers that are incorrect but stated with confidence as if correct. The reasons for this aren’t entirely known. For example, when asking an AI chatbot, “When did Leonardo da Vinci paint the Mona Lisa?” it may respond with an incorrect statement saying, “Leonardo da Vinci painted the Mona Lisa in 1815,” which is 300 years after it was actually painted.
Knowledge Base: A Knowledge base refers to a structured repository of information, such as a database, that provides context and background for language models like ChatGPT. This knowledge base plays a vital role in training AI tools. For instance, language models can be trained using a large collection of text from sources such as books, articles, and scientific journals.
Large Language Model, or LLM: It is similar to a knowledge base, utilising Deep Learning techniques with massive datasets of text and images to enable AI to learn, communicate, and make decisions. When a question is posed, LLMs extrapolate from the available data and teach AI human languages. This allows for the creation of compelling content, as LLMs can even replicate the user’s tone of voice.
Machine Learning, or ML: A component in AI that allows computers to learn and make better predictive outcomes without explicit programming. Can be coupled with training sets to generate new content.
Microsoft Bing: A search engine by Microsoft that can now use the technology powering ChatGPT to give AI-powered search results. It’s similar to Google Bard in being connected to the Internet.
Multimodal AI: A type of AI that can process multiple types of inputs, including text, images, videos and speech.
Natural Language Processing: NLP is an interdisciplinary field encompassing linguistics, computer science, and Artificial Intelligence. Its primary focus is to enable computers to comprehend, analyse, and generate human language. Why is it necessary? Because computers, despite their intelligence, tend to think in a rigid manner. Computers operate in a literal fashion, which poses challenges in processing the nuanced and multiple semantic meanings of certain words.
Neural network: A computational model that resembles the structure of the human brain and is meant to recognise patterns in data. Consists of interconnected nodes, or neurons, that can recognise patterns and learn over time.
Overfitting: Error in Machine Learning where it functions too closely to the training data and may only be able to identify specific examples in said data but not new data.
Parameters: Numerical values that give LLMs structure and behavior, enabling it to make predictions. In the world of computing and AI, a parameter refers to a variable that requires a specific value during programme execution. It helps achieve the desired output.
Prompt: A Prompt serves as an instruction to an algorithm, prompting it to solve a given problem. It acts as a cue for a chatbot to generate a response based on its training and knowledge base. Prompts can take various forms, such as text, code, or even images. However, the quality of your prompt greatly influences the quality of the results you obtain.
Stochastic parrot: An analogy of LLMs which illustrates that the software doesn’t have a larger understanding of meaning behind language or the world around it, regardless of how convincing the output sounds. The phrase refers to how a parrot can mimic human words without understanding the meaning behind them.
Style transfer: The ability to adapt the style of one image to the content of another, allowing an AI to interpret the visual attributes of one image and use it on another. For example, taking the self-portrait of Rembrandt and re-creating it in the style of Picasso.
Temperature: Parameters set to control how random a language model’s output is. Lower temperature values yield more predictable text, while higher values encourage creativity and variation as the model takes more risks.
Text-to-image generation: Creating images based on textual descriptions.
Token: A Token refers to a unit of text that a language model processes. It can be a word, a punctuation mark, or any fragment of text that assists the AI in comprehending the intended requirements. Tokens serve as the fundamental building blocks of text, ranging from individual characters to entire words.
Training data: The datasets used to help AI models learn, including text, images, code or data.
Transformer model: A neural network architecture and Deep Learning model that learns context by tracking relationships in data, like in sentences or parts of images. So, instead of analysing a sentence one word at a time, it can look at the whole sentence and understand the context.
Turing test: Named after famed mathematician and computer scientist Alan Turing, it tests the ability of a machine to behave like a human. The machine passes if a human can’t distinguish the machine’s response from another human.
Weak AI, aka narrow AI: AI that is focused on a particular task and can’t learn beyond its skill set. Most of today’s AI is in fact weak AI.
Zero-shot learning: A test in which a model must complete a task without being given the requisite training data. An example would be recognising a lion while only being trained on tigers.
