Contents
5. Uploading images as part of prompts
AI chatbots: Friends or Foes?
One of my favourite quotes on artificial intelligence (AI). It emphasises the importance of human involvement in the use of AI-based language models and AI-powered machines. Those who master using these tools, gain advantages in the evolving job market, staying competitive and outperforming those who do not.
AI tools like ChatGPT, Gemini, and Copilot are changing how we work, bringing up valid fears of job losses. Sectors that include manufacturing, customer service, and education are the most affected, as the integration of AI in businesses enables repetitive processes to be completed with greater speed, accuracy, and efficiency compared to humans. However, not all jobs are at risk of being taken over. Fields like research, therapy, nursing and design require a high level of innovation, emotional intelligence and/or decision-making abilities that AI-based systems are not yet capable of.
Additionally, the AI revolution itself is creating exciting new opportunities in areas that demand advanced skills, leading to greater job security. These roles are less susceptible to automation because they require uniquely human abilities and expertise. Examples include:
Data Science: The analysis and interpretation of massive datasets to unlock insights that power AI applications. It requires expertise in statistics, machine learning, and programming.
Prompt Engineering: The crafting of prompts (queries) that guide AI models to generate desired outputs. It demands creativity, an understanding of language nuances, and a knowledge of AI algorithms.
AI Ethics Specialism: The development and deployment of ethically responsible AI models, based on fairness, transparency, and societal impact. It requires an understanding of ethical frameworks and AI technology.
AI Chatbot Training: The priming of AI chatbots to understand and respond to customer queries. It necessitates expertise in human-computer interaction, linguistics, and natural language processing.
These are just a few examples, and the landscape is constantly evolving. By embracing ongoing learning and acquiring advanced tech and AI skills, people can position themselves for success in the AI-driven future.
In this article, we will briefly look at how to use Google's AI language model Gemini, even with no previous experience! I will provide easy-to-follow explanations with images provided as a visual aid, alongside example search queries that will showcase how to interact with the chatbot. |
A quick heads up! I'm a big fan of dark mode, so the screenshots you will see below have that setting. When you visit the same pages on your device, they might appear with a white background depending on your own settings. Don't worry, it's the same content, just a different look!
Gemini AI: Google's latest advanced language model
Renamed from Bard AI in early February 2024, Gemini is currently Google's largest and most capable AI language model. It is a great tool to use for research, writing and planning.
The initial Bard chatbot faced some early stumbles with providing factually correct information, even during its launch demo!! In turn, this error highlighted the biggest challenge of using AI chatbots: they make stuff up. This is also one of the primary reasons why human intelligence is required alongside the use of AI: to filter their results, making sure they make sense.
With the relaunch as Gemini, Google's AI chatbot has improved significantly, offering smooth conversations that compete with those of ChatGPT, plus new features that include image generation, and Google Workspace integration. Gemini is now a valuable tool for generating text and answering questions. Applicable for use cases such as writing cover letters, homework assistance, computer coding, detailed translations, and even transcribing voice-to-text.
I have been a consistent user of Gemini since its early inception as Bard. Its accessibility in my African location initially drew me in. Unlike ChatGPT, it was completely free at the time and required no VPN. While we've had our communication bumps, requiring me to refine prompts for optimal results, I prefer Gemini's interface and response generation to other options like Microsoft's Copilot (previously Bing AI), which is run on ChatGPT technology. It has become my daily go-to for refining my writing, drafting emails, and crafting research guides. It's a true timesaver, helping me breeze through tedious tasks and freeing up my time for other endeavours.
Note: For those interested in trying ChatGPT-4 for free or without location restraints, use Microsoft Copilot or other chatbots that integrate GPT-4 technology.
Logging in to Gemini
Go to Gemini's homepage and log in with your Google account. You can use the following link https://gemini.google.com/ or search for Gemini AI on Google. Then log in with an existing Google account or create a new one.
Exploring Gemini: 7 Main Features
Once logged in, a simple and user-friendly chatbot page appears, with 7 primary features that will be discussed below.
Gemini AI Chat Window
1. 'Main menu' & 'New chat'
The Main Menu is a collapsible icon that is available at the top left corner. All previous search chats are saved under the menu, and clicking the 'new chat' button opens a new chat window.
Gemini AI Collapsible Menu
2. Query input area (text, microphone, image)
Prompts (search queries) can be typed into the text area or dictated by clicking the microphone. Images can also be uploaded as part of prompts. For the best results add sufficient detail, in particular, specify:
The writing style - formal, casual, humorous?
Desired usage - blog post, social media caption, product report?
Length - a few sentences, a full paragraph, an article?
Format - script, bulleted list, poem?
The more specific you are, the more Gemini can tailor its response to your needs.
Gemini AI Query Box
3. Main window area and response playback
Exchanges with Gemini are the main focus of the chat window. For each question you ask, Gemini offers 3 unique responses (drafts), presented from different perspectives. You can listen to Gemini's responses instead of reading them by clicking on the speaker icon.
Test Query 1 to Gemini AI: The Initial Prompt
4. Developing the conversation
You can refresh the drafts to see new options, copy them for later use, modify them, provide feedback (thumbs up/down or reporting), share them, or double-check them on Google. However, note that refreshing a search deletes the previous results, and asking a new query only leaves 1 response selection available (rather than all 3 drafts), therefore copy and save any information you may want to retrieve later. Follow-up questions can be asked based on the previous engagement, without a need to restate the original information. Additionally, prior conversations are stored and accessible by scrolling up, until the chat is deleted.
Test Query 2 to Gemini AI: Building The Conversation
5. Uploading images as part of prompts
Images, including screenshots, can be uploaded using the camera icon. The copy (Ctrl-C) and paste (Ctrl-V) commands; (Cmd-C and Cmd-V on Mac), can also be used to insert images directly.
Note: Gemini highlights results that it has double-checked by itself on Google. It also provides links to the relevant pages, which can be accessed by selecting the dropdown icon (arrow) next to the highlighted text.
Test Query 3 to Gemini AI: Using an Image in The Prompt
6. Image generation
Gemini can now be used to generate images, a feature its predecessor Bard did not have. Just like with other AI models, to create images, you simply describe what you want with enough context on the desired look and intended usage.
Test Query 4 to Gemini AI: Creating Images Using Prompts
7. Additional features
Gemini can connect to extensions for YouTube, Maps, Flights, and Hotels (found under "Settings") for more informative replies. It offers usage and privacy support through the "Help" button, as well as personalisation options between dark and light themes under "Settings".
The "Activity" button is used to manage chat history, and as of the recent update, Gemini offers a 'Pro' subscription service similar to other chatbots. It is called 'Gemini Advanced' and is capable of more complex reasoning, planning, and understanding according to Google. This service can be accessed for a monthly fee of around $20. Additionally, as with any other Google service, accounts can be directly managed from the 'user' icon in the top right corner.
Gemini AI Additional Features
Ready to get started with Gemini AI?
Or still feeling apprehensive? I will leave you with a few parting words, that the best way to learn how to swim is by jumping into the deep end. Using Gemini is a lot less scary and not as life-threatening. So don't be shy, experiment, and have fun! There is really no such thing as 'messing up' your queries, and keep in mind that, if your first attempt does not bring the desired result, chances are within 2 - 3 searches you will get what you want.
Through increased engagement with Gemini (or any other AI chatbot), your unique back-and-forth with create a personalised database. Which will be used by Gemini to tailor future responses to your preferences and unique communication style, making each exchange smoother and more relevant.
Take my screenshots for example, notice that I like to keep my prompts simple, direct to the point and conversational. Building up to 'the bigger picture' in a sequential manner, rather than requesting all the details I want at once. I find it helps me regulate the direction of the conversation, refining the responses I get back via streamlined questioning. In time, you can also develop your own techniques for querying and interacting with Gemini, building up to be a prompt engineering pro!
I want to hear from you! Share your favourite AI tools, and how you use them in your work or school!
Stay up-to-date with the latest releases. Join the Newsletter Today!
Comments