Categories
Thinking

Quantum computing

Quantum computing is a new type of computing that uses the rules of quantum physics to process information in ways that traditional computers can’t.

Classical vs. quantum computers

  • Classical computers (laptops, phones) use bits that are either 0 or 1.
  • Quantum computers use qubits, which can be 0, 1, or both at the same time (a property called superposition).

Key ideas (in simple terms)

Superposition: A qubit can represent multiple possibilities at once, allowing many calculations to be explored simultaneously. To understand superposition in a simple way, let’s imagine a person on a ladder [ref. 1]; a person further up the ladder would have a higher potential energy than if they were much closer to the ground. However, unlike the person on a ladder example, atoms can possess more than one energy state simultaneously so the atom would behave like a person who is both occupying the ladder closer to the ground and also further up the ladder. Hence, the atom that is in this mixed energy state is known as “quantum superposition”. [see ref. 1]

Fig 1 – shows superposition in the qubit particle (represented by the sphere) [source ref. 1]

Entanglement: Qubits can become linked so that changing one instantly affects another, even if they’re far apart like if one qubit was placed on the Moon and another qubit was placed on Earth [source ref. 1]

Fig 2 – shows quantum entanglement between two qubit particles (each qubit particle represented by a sphere) [see ref. 1]

Quantum interference: The computer amplifies correct answers and cancels out wrong ones through carefully designed operations. Interference arises because of the wave-like properties of quantum particles like electrons and photons (a photon being a packet of energy of electromagnetic radiation, [see ref. 2 and 3]). When a particle is in a superposition of multiple states, these states can interact with each other that can lead to constructive or destructive interference [see ref. 2].

Fig 3 – shows constructive wave interference on the left and destructive wave interference on the right [see ref. 4]

Why this matters

Because of these properties, quantum computers can solve certain problems much faster than classical computers, such as:

  • Breaking or analyzing some types of encryption
  • Simulating molecules and materials (useful in medicine and chemistry)
  • Optimizing complex systems (like logistics or traffic flow)

Important limitations

  • Quantum computers are not faster at everything.
  • They are very hard to build and control.
  • Current quantum computers are experimental and prone to errors.

Simple analogy

If a classical computer checks solutions one at a time, a quantum computer explores many paths at once, then uses physics to guide itself toward the best answer.

Bottom line

Quantum computing doesn’t replace regular computers. Instead, it’s a powerful new tool designed to tackle specific problems that are extremely difficult or impossible for today’s machines.

Reference:

  1. https://www.nist.gov/quantum-information-science/quantum-computing-explained
  2. https://quantum.microsoft.com/en-us/insights/education/concepts/interference
  3. https://www.britannica.com/science/light
  4. https://kids.britannica.com/students/assembly/view/53869

Categories
Thinking

What is an AI chatbot?

An AI chat bot is a computer program that simulates human conversations. An AI chat bot uses AI to understand and respond to queries. An AI chat bot uses AI techniques such as natural language processing (NLP) to understand human language. NLP is the processing of natural language information by a computer via understanding, analysing and responding to human speech. NLP takes a text or voice from a human, uses AI to make sense of the text or voice, then generates and delivers responses to the text or voice input.  

NLP splits simplifies text by making text into tokens i.e. breaking down sentences into individual parts e.g. the sentence “What restaurants are nearby?” is broken down into the following tokens: ‘what’ ‘restaurants’ ‘are’ ‘nearby’. By converting text into tokens, it removes any less useful information.

AI chat bots use Large Language Models (LLM) to understand and recognize text. Think of LLMs as a type of AI program. LLMs take the sentences that have been made into tokens and then analyse how words connect and which words frequently appear together.

Types of AI chatbots:

The most common types of AI chatbots are: transactional AI chatbots, Conversational chatbots, Decision support AI chatbots, Workflow automation AI chatbots, Informational AI chatbots, Problem-solving AI chatbots, Feedback and survey AI chatbots, Adaptive learning AI chatbots and Hybrid AI chatbots.

Transactional AI chatbots operate on orders, deliveries and transactions.

Conversational AI chatbots are usually used in customer service roles, messaging human users in a conversational way.

Decision support AI chatbots help human users make decisions by presenting data and outcomes to the user.

Workflow automation AI chatbots simplify repetitive tasks in a workflow.

Informational AI chatbots provide information and answers FAQs.

Problem-solving AI chatbots diagnose and answer customer problems.

Feedback and survey AI chatbots gather data from customer surveys and feedback and develop insights based on the data.

Adaptive learning AI chatbots improve their performance based on the feedback they receive and the improvements they implement.

Hybrid AI chatbots can switch between transactional, informational and problem-solving roles for human users.

Categories
Thinking

Understanding AI Chatbots: How They Work and Types Explained

What is an AI chatbot:

An AI chat bot is a computer program that simulates human conversations. An AI chat bot uses AI to understand and respond to queries. An AI chat bot uses AI techniques such as natural language processing (NLP) to understand human language. NLP is the processing of natural language information by a computer via understanding, analysing and responding to human speech. NLP takes a text or voice from a human, uses AI to make sense of the text or voice, then generates and delivers responses to the text or voice input.  

NLP splits simplifies text by making text into tokens i.e. breaking down sentences into individual parts e.g. the sentence “What restaurants are nearby?” is broken down into the following tokens: ‘what’ ‘restaurants’ ‘are’ ‘nearby’. By converting text into tokens, it removes any less useful information.

How AI chatbots work:

AI chat bots use Large Language Models (LLM) to understand and recognize text. Think of LLMs as a type of AI program. LLMs take the sentences that have been made into tokens and then analyse how words connect and which words frequently appear together.

Types of AI chatbots:

The most common types of AI chatbots are: transactional AI chatbots, Conversational chatbots, Decision support AI chatbots, Workflow automation AI chatbots, Informational AI chatbots, Problem-solving AI chatbots, Feedback and survey AI chatbots, Adaptive learning AI chatbots and Hybrid AI chatbots.

Real world applications:

Transactional AI chatbots operate on orders, deliveries and transactions.

Conversational AI chatbots are usually used in customer service roles, messaging human users in a conversational way.

Decision support AI chatbots help human users make decisions by presenting data and outcomes to the user.

Workflow automation AI chatbots simplify repetitive tasks in a workflow.

Informational AI chatbots provide information and answers FAQs.

Problem-solving AI chatbots diagnose and answer customer problems.

Feedback and survey AI chatbots gather data from customer surveys and feedback and develop insights based on the data.

Adaptive learning AI chatbots improve their performance based on the feedback they receive and the improvements they implement.

Hybrid AI chatbots can switch between transactional, informational and problem-solving roles for human users.

Advantages and limitations:

Advantages of AI Chatbots

1. Available 24/7

  • AI chatbots can answer questions at any time without needing breaks.
  • Businesses can provide customer support outside normal working hours.

2. Fast Responses

  • They can respond within seconds, reducing waiting times for users.

3. Cost-effective

  • Organisations can automate answers to frequently asked questions, allowing human staff to focus on more complex issues.

4. Consistent Information

  • A chatbot gives the same answer each time when asked the same question, helping maintain consistency.

5. Scalable

  • One chatbot can handle thousands of conversations simultaneously, whereas a human customer service team has limited capacity.

Limitations of AI Chatbots

1. Lack of Human Understanding

  • Chatbots may struggle to understand sarcasm, humour, emotions or ambiguous questions.

2. Can Produce Incorrect Information

  • Some AI chatbots may generate inaccurate or misleading responses, so users should verify important information.

3. Limited Reasoning in Some Systems

  • Rule-based chatbots can only answer questions they have been programmed to recognise.

4. Privacy Concerns

  • Users should avoid sharing sensitive personal or financial information unless they trust the platform and understand its privacy policy.

5. Dependence on Training Data

AI systems are only as good as the data they have been trained on and the models that power them.

Future Developments

More Natural Conversations

Future chatbots are expected to become better at:

  • understanding context,
  • remembering previous parts of a conversation,
  • producing more natural responses.

Voice and Multimodal AI

Rather than relying only on text, AI chatbots increasingly combine:

  • speech recognition,
  • image understanding,
  • document analysis,
  • video interpretation.

Users may be able to show a chatbot a photograph, upload a PDF or speak naturally instead of typing.

Personalised Assistance

Future chatbots may provide more personalised recommendations based on a user’s preferences and previous interactions, while respecting privacy settings.

Education

AI chatbots are likely to play a growing role in education by:

  • explaining difficult concepts,
  • generating practice questions,
  • providing personalised feedback,
  • supporting independent learning.

As a science tutor, this is a topic you can discuss from first-hand experience.

Healthcare

Healthcare organisations are exploring AI chatbots to:

  • answer common health questions,
  • remind patients to take medication,
  • assist with appointment scheduling.

However, they are generally intended to support—not replace—qualified healthcare professionals.

Business Automation

Businesses are increasingly integrating chatbots with:

  • customer service,
  • online shopping,
  • booking systems,
  • internal knowledge bases.

This can improve efficiency and reduce response times.

Ethical and Regulatory Challenges

As AI becomes more capable, important questions include:

  • protecting user privacy,
  • reducing bias,
  • ensuring transparency,
  • preventing misinformation,
  • defining appropriate regulation.

Conclusions

AI chatbots have transformed the way people access information and interact with technology. While they offer significant benefits in education, business and customer support, they are not infallible. Users should think critically about the information they receive, verify important facts using reliable sources and recognise that AI is a tool to assist human decision-making rather than replace it.