🔥 Weekly AI news: Did you miss it?!

👨‍🚀 The most important AI updates at a glance

AI-HOI, AInauts!

Maybe you didn't catch all the news, tools, and hacks about AI last week, or maybe you've only recently joined us. Either way, here's our recap with all the headlines from the newsletter - just one click away!

Click the links to jump right to the article - or read our picks below.

• 🎢 OpenAI's master plan: The road to artificial superintelligence

• 🌐 Claude Artifacts: How to share and remix content!

• 🤩 AI video roundup: These AI clips are crazy!

• 💻 How companies are turning their employees into AI users

• ✍️ This prompt makes your marketing copy better

• 📼 Viral AI video: Stars and their younger selves

→ Selection of the top posts of the last week ←

🎢 OpenAI's master plan: The Road to Artificial Superintelligence - Opportunity or Risk?

Imagine you wake up tomorrow and your smartphone has developed an IQ of 160 overnight. That will happen - the only question is whether in 6 months or 2 years ...

An important driver behind this timeline is OpenAI. The company presented its ambitious roadmap internally a few days ago. And what we see there is nothing less than an attempt to redefine the boundaries between man and machine.

It is therefore time to put these plans under the microscope.

🚦 OpenAI's 5-step plan to superintelligence

In an internal all-hands meeting, OpenAI presented a 5-step scale that describes the path to Artificial General Intelligence (AGI). Sounds dry at first?

Wait and see, because this scale is the key to understanding where the journey is going:

1. 🟢 Chatbots(We are here!): These are our current assistants like ChatGPT, Claude, etc. Clever, but still a long way from real comprehension.

2. 🟡 Reasoners: The next level. These systems solve problems at the level of a human with a doctorate title - and without external tools. OpenAI claims to be on the verge of this breakthrough.

3. 🟠 Agents: Imagine AI assistants that can carry out complex tasks independently for days on end. Your personal digital assistant, so to speak. And of course, with a doctorate title in all disciplines.

4. 🔴 Innovators: This is where it gets really exciting (or scary, depending on your point of view). AI systems that make new discoveries, invent things on their own and achieve scientific breakthroughs.

5. Organizations: Completely AI-driven companies that could theoretically replace an entire organization.

Will these AIs develop a consciousness at some point? Some experts believe that we will soon be having heated debates about this. But that’s for another day.

In any case, OpenAI CEO Sam Altman believes that we could reach AGI before the end of this decade. But how are they going to do it? This is where "Project Strawberry" comes into play.

🍓 Project "Strawberry": OpenAI's secret future technology

According to an exclusive Reuters report, OpenAI is working on this secret technology (remember Q*?) to take the capabilities of AI models to a whole new level.

What makes Strawberry so special? It is designed to enable AI agents to:

• Roam autonomously on the Internet

• Plan ahead

• Carry out "in-depth research"

Imagine giving AI the task of solving a complex scientific problem. Instead of just processing existing information, it could:

1. Find and analyze relevant research papers

2. Simulate experiments

3. Collaborate with other (AI) systems

4. Formulate and test new hypotheses

And we're talking about multimodal agents that can inherently understand and generate text, code, image, audio and video (text, vision, voice).

But wait a minute... If the agent is so autonomous, who controls what it does and learns there? We’ll find out, but this seems to bei OpenAI's vision for the near future.

Will OpenAI stun us with a new release soon? On the one hand, a new, improved GPT mini model has appeared in the chatbot arena - and on the other hand, Chinese VPNs have been blocked from the GPT models for a few days now. China itself had already censored ChatGPT behind the Great Firewall, but now OpenAI is also pulling the plug on their end as well. Something is going on ...

Anyway, if you look at the development of use cases from GPT-3 to GPT-4, it no longer takes much imagination to make the above-mentioned use cases a reality.

🕵️Whistleblowers and regulatory authorities sound the alarm

Plot twist! While OpenAI raves about a bright future, some employees are sounding the alarm. These whistleblowers have now filed a complaint against OpenAI with the US Securities and Exchange Commission (SEC).

The accusation? Employees were prevented from disclosing potential security risks of their technology by restrictive confidentiality agreements. Employees had to sign contracts that prevented them from disclosing information to the authorities.

Security is an ongoing issue. OpenAI has also just announced a collaboration with the Los Alamos National Laboratory (which has also developed the atomic bomb) to research the safe use of AI in the laboratory environment, particularly in the life sciences. The mandate behind this comes from the White House.

And in the midst of all these developments, Microsoft - OpenAI's main investor with a 49% share - recently announced that it would give up its seat on the OpenAI board. Instead, there will be bilateral meetings. They have seen "significant progress" over the last eight months and are confident about the direction of the company.

But what is really behind this retreat? Is it a sign of trust or strategic distancing? Given the increasing attention from antitrust authorities to the links between tech giants, this is a clear sign of pre-empting regulatory concerns (not least because of the EU Competition Commission).

And Apple also wants to give up the promised board seat for this reason …

🎯 Where do we really stand? A critical assessment

Phew, part of us can hardly wait to see how these technologies could change our lives and our work. On the other hand, we have to admit: The speed and scale of these developments are also a little frightening.

The ethical questions, the security concerns, the coming social upheavals - all of this sometimes makes us uneasy in quiet moments.

But despite all our concerns, we remain cautiously optimistic. Why? Because we see that a critical discourse is taking place. Whistleblowers have the courage to expose abuses. Big players are working together to investigate the risks.

The path to AGI seems chaotic and unpredictable - but technological progress has always been that way. What we want to say is this: Yes, we should remain vigilant and question critically - but not lose sight of the fascination and opportunities that this development brings with it.

Either way, we are living in the most exciting time that is setting the course for the future.

🌐 Claude Artifacts: How to share and remix content!

Anthropic has set the bar high with Claude 3.5 Sonnet: it's the best chatbot with projects and artifacts. This new feature allows you to create extensive, independent text documents, code snippets or interactive HTML pages - and now you can also publish, share, remix and further develop them.

Interactive content and use cases

Content and ideas can easily be transformed into apps, games, quizzes, simulations, etc. - playful and interactive. We have clicked through dozens of examples, here are the 20 best.

Games like Whack a Mole, Worm Battle, Tetris, Triple Tic Tac Toe, Pacman, Rubics Cube, Fireworks, etc.

Apps like To Do Listen, AI Video Prompt Generator, Investment Calculator, SVG Pattern Builder, Image Processor, Fake Food Delivery, …

Learn how to Prompt, to Scrape, Quantum Mechanics, Flash Cards to learn AI, Climate Change, Mindmaps, …

More in the Claude Artifacts Directory or Claude Artifacts Showcase (unofficial). Click through and try it out!

P.S.: The popular chatbot Poe has also rolled out a similar feature.

🤩AI video roundup: These AI clips are crazy!

The developments in the field of AI videos over the last few weeks have been intense; we have already reported several times on Runway, Dream Machine, KLING and Sora.

To give you a quick overview of what is already possible, we have put together the best clips for you - enjoy!🔒 AI scraping: How to protect your website from bots

In order for our beloved AI models such as ChatGPT, Claude & Co. to become really smart, they need to be fed with data and knowledge. From the internet, of course. And how? By sending out an army of small robots or bots, surfing the web, sucking up and storing the content of all websites.

You’ve heard it before; the whole thing is called web scraping or crawling.

So far so good. Perhaps you can imagine that many website owners don't like this. And that's also a reason why there are various lawsuits against AI companies, often brought by media companies.

How to protect your website from AI bots and scrapers

If you don't want your website to be crawled by scrapers, you have to do something about it yourself. And there are many of these little bots, as you can see in the graph below:

As you can see, the biggest data collectors are ByteSpider (by the Chinese company ByteDance, which also owns TikTok), Amazon-Bot, Claude-Bot and GPT-Bot.

The problem is often that only a few large providers - OpenAI, Google, etc. - label their bots properly. But there are also countless other bots that pretend to be just normal web browsers.

Currently, the easiest way to protect your website from these bots is to use a service called Cloudflare.

Cloudflare is a provider of website security and also offers faster loading times. It already provides very good protection against various hacker attacks, such as DDoS attacks etc.

Cloudflare now also allows you to block all AI bots with one single click. This is based on a proprietary machine learning model that also blocks the bots that pretend to be regular web browsers:

Bots blocked: Will all AI models now remain up to date?

As website operators, we naturally like the Cloudflare service.

As AI model users, not so much. Cloudflare is a really widespread service, super easy to implement and already 80% of the surveyed Cloudflare users want to block bots with it.

And as a result, the new models might lack new data.

It remains to be seen how big the impact will really be. There will probably be a cat-and-mouse game between scrapers and website operators.

In addition, many AI providers are already licensing content from media companies and online platforms. OpenAI, for example, has agreements with Axel Springer, TIME, Reddit, Vox and many others.

Or you can do it like Microsoft, which simply bought the GitHub platform for a few billion back in 2018...

💻 How companies are turning their employees into AI users (in 15 minutes)

Every day we deal with companies that want to implement AI into their organization. This often involves introducing individual teams to the possibilities that AI brings to the table and show them how to use it in their workflows.

The aim is for each individual employee to get to know, like it and ideally use an AI assistant to produce better results.

In any case, the most important thing is to understand the basics.

Many people have already played around with ChatGPT, but unfortunately, they have often produced only mediocre results. And then the motivation quickly fades.

So, here's a 15-minute exercise from us to quickly get people excited about AI - and achieve very good results in the process.

Step 1: The best shortcut for prompting

Mediocre results are a killer - and the reason why many people stop using AI at all.

But this mediocrity usually stems from insufficient prompting efforts. And that's why we always spend some time showing others how to come up with a good prompt that delivers the desired results.

A good prompting strategy includes:

• Context & specificity

• Roles & "hats"

• Step-by-step instructions

• Examples (Good and bad)

Most people are put off because it takes time and testing to create good prompts.

To address this, the first thing we show them is the Anthropic prompt builder. It turns a prompt like this ...

Write me attention-grabbing LinkedIn posts based on topics I specify.

… info a prompt like that - with one single click!

Extremely helpful, and this virtually solves the issue of "the more time spent on a prompt, the better the result".

Step 2: Get to know the magic and deepen your knowledge

The second step is to stay with Anthropic and use their chatbot Claude. You can easily create a free account to get your feet wet.

The second goal we have is to show people a little magic by letting them build a small web app.

Here is a prompt example:

Please create an interactive game for me to learn what the most important AI prompting techniques are and why prompting is so important. 

Please include at least 6 prompting techniques and at the end a task to write a prompt.

Thanks to the Claude Artifacts, the AI then gets started and builds a small app that explains prompting and even has a final mini-exam.

And thanks to the new Publish-function, participants can also bring their own app online and even share it with others.

Here is the first result from a two-line prompt.

With these two simple steps and just one free tool, you can get and motivate people to use AI in 15 minutes and arm them with valuable knowledge most people don’t know about! It's easy, and you are not overwhelming anyone with 1000 different tools and possibilities.

✍️ This prompt makes your marketing texts better

We are often writing about prompts that improve our results. That's why we're once again sharing a prompt for anyone in the marketing field.

But you can also use it for all kinds of content where you want people to take action.

The prompt uses the emotion framework of the well-known podcaster Shaan Puri.

A while back we’ve participated in his online course on copywriting, which was really, really good. (Unfortunately, Shaan is now extremely rich and no longer offers the course).

Anyway, here's some of his best content packaged into a prompt you can use right away.

Before we dive delve in, a quick background story: We can reach people better if we appeal to their emotions.

And: The right emotions are the difference between boring and viral.

What emotions are we talking about? Shaan has developed a cool framework, which is especially designed to get people’s attention.

He describes the following emotional states:

• LOL: that is so funny

• OHHH: now I understand!

• WOW: that is amazing!

• AWW: this is sooo cute

• YAY: that's great news!

• WTF: that makes me angry

• FINALLY: finally someone says it!

We want to trigger at least one of these emotions with our content. This is where the following prompt comes into play:

You have to analyze a text and make it more interesting with Shaan Puri's emotion framework. The aim is to make the text more viral. This is how you proceed:

Step 1: Read the following text carefully: 
<text_sample>
[Insert your text here] 
</text_sample>

Step 2: Analyze the content and think about how you could adapt or improve it to trigger these feelings:

* LOL: that is so funny
* OHHH: now I understand!
* WOW: that's amazing!
* AWW: this is sooo cute
* YAY: this is great news!
* WTF: that makes me angryd
* FINALLY: Finally someone says it!

Step 3: For each feeling, give a concrete suggestion on how the text could be adapted to evoke that exact feeling. Your suggestions should be practical, feasible and in keeping with the original text.

Present your analysis and suggestions in this format with markdown headings:

Feeling: [Feeling]
1 [First proposal]

Make sure that your suggestions:

* Specific and tailored to the content of the text are
* Maintaining the original intention and message of the text
* Realistic and feasible are
* Do not change the core content completely, but rather improve or adapt it

If there is a feeling that is particularly difficult to incorporate, you can mention this briefly. Try to make creative suggestions anyway.

Remember, your goal is to make the text more engaging and shareable by appealing to these emotional responses. Be creative, but be realistic in your suggestions.

Now you just add some context or your existing text between the <text_sample> tags and you get an emotionalized version of it.

Pretty cool, isn't it?

Your AInauts, Fabian & Reto

Follow us on Twitter & LinkedIn!

Your feedback is essential for us. We read EVERY comment and feedback, just respond to this email. Tell us what was (not) good and what is interesting for YOU.