It’s Friday again…

Here’s The Weekend Express:

• How to use ChatGPT one year on

• Say goodbye to your job and hello to AI

• Our AI Tool of the Week - Social Media

• How to stop ChatGPT from lying to you

• Weekend LOL’s

📰 Read time = 4 mins

How to use ChatGPT one year on from inception

Next month will signify one year since the mainstream usage of ChatGPT. So what have we learnt so far?

Workers in the corporate world have used ChatGPT as an assistant copywriter, a jargon demystifier, and even asked the AI bot how to sound ‘less like an ass-hole’ in emails.

In juggling the everyday life tasks of us mere humanoids we have turned to ChatGPT as our Sous Chef, asking what we should have for dinner tonight and ‘what can I make with these ingredients.’

With last months addition to ChatGPT of the ‘image interpreter’ (have you used it yet, it’s amazing!) users are now asking it to create plans from whiteboard notes. Ridiculous feature!

Read more on The Guardian

What have you used ChatGPT for in the first year or are we still just creating LinkedIn posts? Explore and let us know.

Say goodbye to your job, and hello to your new AI job

We’ve heard it time and time again… AI is going to take my job. Maybe, but there’s more…

In a recent fearful statement by Morgan Stanley they predicted that as many as 300 million jobs could be at stake with the introduction of AI.

But fear not, your use is not redundant just yet. It is said that a new wave of AI reviewers and assisters will be introduced to hold the hand of the AI bots. Does that mean our roles are reversed of how we see them now?

Phelim Bradley, CEO of prolific, said that there are plenty of new kinds of “AI workers” who are playing a key role in informing the data that goes into AI models like ChatGPT — and what comes out. A new job description requirement on job ads is born. - Must be able to work with AI

Full story at CNBC

Nervous about losing your job and don’t know where to start to become an AI reviewer. Don’t worry, you’re in the right newsletter. 😄 

Our AI Tool of the Week (which isn’t ChatGPT)

It’s hard to keep content thriving across all social media channels. Instagram, Youtube, Tiktok, Snapchat, what else?

Well now you can generate multiple social media content from long form videos.

You can turn any of your long videos or youtube content into 10 or more Tiktok videos or Instagram reels, at the click of a buttom.

Opus Clip will also generate speech to text and rank your videos by it’s ‘viral capabilities. Don’t believe us just try for yourself.

Get out there and produce that social media content, easier than ever before.

ChatGPT keeps lying to you - How to make it stop

Have you ever prompted ChatGPT and had it knowingly lie to you? Well sometimes even ChatGPT gets mixed up. We’ve even had the master bot apologise mid reply and start a new reply. You can also question the response of your ChatGPT replies by asking it to confirm it’s all true. It’s worth your time.

Here are 5 ways to make sure ChatGPT is not lying to you. (Knowingly or unknowingly)

1. Use simple, direct language

One of the main causes of hallucination is ambiguity. When you use complex or vague prompts, the AI model may not understand what you want or what you mean. It may try to guess or fill in the gaps, resulting in inaccurate or irrelevant responses.

2. Incorporate context into your prompts

Another way to reduce ambiguity is to provide some context in your prompts. Context helps the AI model narrow down the possible outcomes and generate a more relevant and appropriate response. Context can include information such as your location, preferences, goals, or background.

3. Give the AI a specific role – and tell it not to lie

Sometimes, AI can make up stuff when it does not have a clear sense of its identity or purpose. It may try to imitate human behaviour or personality, which can lead to errors or inconsistencies. It may also try to ‘impress’ you by making up things that are not true or realistic.

4. Limit the possible outcomes

Another reason why hallucination happens is because the AI model has too many options or possibilities to choose from. It may generate something that is random or unrelated to your prompt. It may also generate something that is contradictory or inconsistent with previous responses.

5. Pack in relevant data and sources unique to you

Finally, one of the best ways to prevent a chatbot from spewing misinformation is to provide relevant data and sources unique to you in your prompts. Data and sources can include facts, statistics, evidence, or references that support your prompt or question. Data and sources can also include personal information or experiences that make your prompt more specific or unique.

Read more at Indian Express

Like any tool it is only useful if you know how to use it, try these tips for yourself to hone your ChatGPT skills. ✍🏻 🥷

A word from this week’s sponsors…

BIT Training provides award-winning training in IT and Cyber Security.

We provide flexible learning solutions whenever you need us. Visit us now to learn more.

Weekend LOLS

Here’s a visual representation of our article in this email about ChatGPT lying to us. What a predicament to be a bot.

Have you asked ChatGPT not to lie to you?

Let us know how you get on and we’ll see you Monday, have a great weekend buzzers.

Send in your memes to [email protected]