In this weeks issue, I complete my look at how the cities of the near future might look and function. This is slightly more crystal ball gazing, but all based on threads that you can pull right now. I hope you enjoy it. In the last of the series of Remember When, I’ll be looking at the computer, or at least the earliest version of it. I started this series with the telephone and have gone through some of the things that I can remember and anecdotes from them, but I think it has run its course. I’ve enjoyed writing the series but next week will see a new look and unfortunately, there is no place for the nostalgia.

News

Dall-E 3 to be integrated into ChatGPT: OpenAI have announced that their latest version of the text to image editor, Dall-E 3, will be integrated into ChatGPT from October, allowing same text editing to create fantastic images. This ease of use represents a huge challenge to market leader in this space, MidJourney. Get the scoop here

Big AI leap forward in predicting disease: Google DeepMind has developed AlphaMissense, an AI tool that could transform how we understand genetic mutations linked to diseases. Built on their groundbreaking AlphaFold technology, this new tool offers more precise predictions on which protein mutations could result in health conditions. Read all about it here

UK Deputy PM Sounds Alarm on AI's Impact at UN: UK Deputy Prime Minister Oliver Dowden has urgently called for international regulation of AI in a speech at the UN General Assembly. Highlighting fears like job loss and the spread of misinformation, Dowden announced that the UK will host a global summit in November to tackle these pressing issues. Experts weigh in, noting that the rapid advance of AI technology poses challenges that demand immediate, collective action. Get the lowdown here

Is Europe killing itself financially with AI regulation: The proposed AI Act in Europe is causing divisions among experts and industry leaders. While some view regulation as essential for safety and ethical concerns, others worry it could stifle innovation and economic prosperity. Read more here

ChatGPT will soon be able to engage via voice and image: OpenAI is enhancing ChatGPT with voice and image search functionalities, allowing for more interactive user experiences. While the voice feature aims for better accuracy than competitors like Alexa, and a new text-to-speech model offers "human-like audio," OpenAI is cautious about the ethical implications, including impersonation risks. The image search function also brings forth questions of privacy and accuracy. Interesting times ahead! Read here

Just One Thing…

Following on from part 1 of my discourse on the city of the future, I wanted to dream a little. I had originally thought that I would look out to 2030 but have decided to call it 2050 as I, at least, am still getting used to AI based timeframes and can’t imagine how things could change so quickly having spent a lifetime in the placemaking business where things move at a snails pace. The important thing to note here is that none of the things I try to articulate here are far out. All of them, in one way or another, are being tested or prototyped right now. So this could be 2030. But I think, more like 2050!

Imagine a city that thrives in harmony with nature rather than exploiting it—a place where the built environment complements rather than competes with its natural surroundings. The city of the near future could be a blueprint for harmonious, sustainable living, fully integrated with ecological intelligence. This vision leans on multiple components: from sustainable land use and energy sourcing to socially inclusive infrastructure and resourceful waste management, all with AI at it’s beating heart.

Sustainable Land Use

In a densely populated urban core, sustainable land use emerges as a cornerstone for survival. This implies more than just consents and permissions; it is about holistically understanding how the city can provide essential resources like water, food, and recreational spaces. In the city of the near future, high-capacity transit options would play a crucial role in not just reducing emissions but also in speeding up commute times.

Ecology: In Harmony with Nature

The design philosophy of future cities will surely revolve around aligning with natural features and forces. This is the direction of travel we see right now and I can see no returning from this. Wildlife habitats are protected, and natural resources are conserved, all orchestrated under a unified vision for the entire region. One innovation in this realm could be bioswales or absorbent rain gardens that replace conventional gutters, capturing and filtering rainwater for reuse.

Water: The Lifeblood

Protecting upland water systems and rigorous stormwater management measures are not just about human consumption but also about protecting aquatic habitats. "Sponge city" measures already tested in places like Shanghai allow water to percolate through soil, recharging water tables and preventing flooding. Managing water risk in this way is essential going forward.

Social and Sustainable Transit

Transit stations in this future city aren't just transport hubs; they're centres for business and social activities. High-speed rail options could make travel not just faster but also significantly more sustainable. Mass transit systems that are clean, safe and central are so important when designing the city of the future

Renewable Energy: Self-Sufficient Urbanites

Solar panels (on walls and roofs) and roof gardens would become commonplace, leading to a city that generates at least as much energy as it consumes. In fact, the entire energy matrix would be 100% renewable. Area buildings could share energy resources, converting waste into energy.

Waste Not, Want Not

The city of the near future will aim to transform waste into a resource. Soil remediation initiatives would reclaim landfills and abandoned industrial areas, while wastewater would be treated for irrigation and even human consumption. Moves are afoot in this area all around the world already.

Smart Buildings: Designing to Scale

Buildings in this city would be modular, adaptable and incorporate natural elements, making construction faster and waste minimal. Such spaces can quickly transform to meet changing housing, industrial, or business needs. The buildings themselves would be smart, imbuded with sensors, detectors, cameras and multiple other devices that can ensure that the building runs as efficiently and safely as possible.

Food: From Sky Gardens to Local Produce

Urban farming would benefit from advanced hydroponic technology. Global standards for organic farming and animal treatment would be established, and most produce would be sourced locally, reducing the carbon footprint of food transportation. The concept of minimising energy use from farm to plate will prevail in the coming years.

Mobility and Livability

The rise of automated technology and high-speed rail would make future cities more navigable and safe. Pedestrian friendly designs, reduced car usage, and more plants both inside and outside buildings would improve air quality.

Culture and Social Interiors

Historical heritage would be preserved and celebrated, and virtual and augmented reality could bring arts and entertainment to a global audience. Shared spaces and community activities would foster social interaction and equality. Cities at their best, are places where life happens, be it in groups, in clubs, in neighbourhoods, precincts or entire cities. A thriving hub of sports, culture, arts and entertainment.

Economic Resilience

As artificial intelligence and automation permeate society, people would adapt to more flexible working hours. Economic policies would be crafted to be ecologically sustainable, promoting a holistic approach to urban living.

Conclusion

The city of the near future is not just a dream; it is an urgent requirement for a future where people and the planet can coexist and thrive. From energy to social interaction, each aspect should be designed to be sustainable, inclusive, and adaptable to change. With cohesive planning and communal effort, such a city is within our grasp. It merely needs the political will (and therein lies the rub, hence 2050) to make it happen.

Remember When…

In the annals of history, the early 1980s resonate as a period of astounding technological change—a gateway to an era that would redefine our understanding of the world and our place in it. If you were fortunate enough to be around at that time, you’ll no doubt remember this:

The Sinclair ZX81. With its basic 1KB memory and modest processor, this tiny marvel could fit on your desk and had the power to introduce us to a whole new digital universe. It was rudimentary, but it was revolutionary.

Fast forward a few decades, and here we are, residing in a reality where machines have transcended our wildest imaginations. Today, supercomputers powered by awe-inspiring GPUs harness the prowess of trillions of transistors, executing calculations at speeds that would've boggled the 1980s mind. It's as if we've gone from building sandcastles with toy spades to designing architectural marvels with state-of-the-art cranes and 3D printers.

The Nostalgic '80s

Let's jump back to the 1980s for a moment. Schools were just introducing computers, and as a senior pupil, I had the privilege of booking a half-hour slot to interact with these digital wonders every week. I'd arrive to the lab and retrieve my very own 5.25-inch floppy disk and insert it with a mixture of trepidation and excitement. There was a programme at the time that allowed you to simulate macroeconomic models—input your variables, economic growth, unemployment, inflation, etc and hit “RUN”. The problem was, you had to wait a whole week to find out if your model had ushered in prosperity or fiscal disaster!

The Predictable Future 

Perhaps what's most astonishing is that, even then, the future was written in the silicon. I was introduced to Moore's Law at that time: the prediction that the number of transistors on a microchip would double approximately every two years, thereby massively increasing its computational power. It was almost like a divine technological scripture. In economics class we were tasked with calculating the potential of transistors in the year 2000 and beyond. Could anyone then fathom the marvels of technology that would ensue?

The AI Epoch

And here we are, living in the times foretold. Modern chips are currently packed with up to 1.7 trillion transistors. This astronomical leap has enabled the rise of AI, which is penetrating every facet of life. The algorithms of today have an analytical depth that dwarfs anything the Sinclair ZX81 could've even feebly attempted. From machine learning applications that predict economic trends with remarkable accuracy to natural language processors that can write entire books, we've entered an era where computational power is, for all practical purposes, limitless.

The Unfathomable Future

Imagine what the world will look like in two years, given the current exponential growth. As AI algorithms become increasingly complex and efficient, we can expect a surge in capabilities. Moore's Law may face physical limitations, but innovations in quantum computing promise to pick up the mantle. It's not just about transistors anymore; it's about reimagining the very architecture of computation itself.

Conclusion

From the humble beginnings with the Sinclair ZX81 to the technological titans of today, our journey has been nothing short of miraculous. The nostalgia of waiting a week to know if your economic model succeeded is replaced by instant gratification and sophisticated analysis at the tap of a screen. Yet, the spirit of discovery, the awe of what's possible, remains unchanged. We were always headed here, each transistor a stepping stone to our unfathomable future. AI isn't just an application of this computational power; it's the embodiment of our collective curiosity and ambition. The question is no longer "what can these machines do?" but rather "what can we accomplish together?"

Practical Prompts…

This week, I want to show you an example of how I use ChatGPT for brainstorming ideas and content. Here goes:

Me: I want to brainstorm with you some prompts that I can use as tweets to get the best out of ChatGPT. Is that something that you can do? I'm looking for actionable content that my audience can use straight away.

ChatGPT: Certainly, brainstorming prompts for tweets that can help you engage your audience in actionable ways is something I can assist with. Since you're interested in AI and its implications on various aspects of life, I'll focus the prompts around that theme.

1. "What's the single most transformative use of AI you've seen in property development? Comment below."

2. "How do you see AI shaping the future of sports analytics? Let's discuss the potentials and pitfalls."

ChatGPT then went on to list another 18 tweets.

Me: Okay, I wasn't thinking about that exactly, but they are certainly useful. Let me try and be more specific. I want to be able to share principles for getting the best output from ChatGPT (you!). In other words, how should I structure what I ask you to make sure that you understand what I am asking. Does that make sense?

ChatGPT: Absolutely, your request makes perfect sense. If your goal is to share insights on how to get the best output from me, then focusing on the principles of query formulation can be incredibly beneficial. Here are some principles that your audience might find actionable:

1. Be Specific: Encourage people to be as specific as possible in their questions. For instance, instead of asking, "Tell me about AI," a more specific query would be, "How is AI used in property development?"

2. Clarify the Context: If a question could have multiple interpretations, providing context can lead to more accurate answers. Example: "In the context of macroeconomics, how does AI impact employment?"

3. Request the Desired Format: If your audience wants a list, summary, or step-by-step explanation, they should state that in the question to guide the output.

4. Word Count Constraints: If they need a brief or lengthy reply, specifying a word count can help tailor the response to their needs.

5. Ask for Pros and Cons: To get a balanced view on a topic, one can ask for both advantages and disadvantages.

6. Ask for Reasoning: Encouraging people to request the rationale behind an answer can yield more insightful responses. Example: "Can you explain the steps you took to arrive at that conclusion?"

7. Prompt for Examples or Analogies: Sometimes, complex topics can be easier to understand through examples or analogies. Explicitly asking for these can be useful.

8. Limit to One Topic per Query: Multiple topics in a single question can result in a diluted or confusing answer.

9. Directly State the Goal: If the goal is to learn, debate, or solve a problem, stating this can help shape the response appropriately.

10. Iterate and Refine: If the first answer isn't satisfactory, encourage them to ask follow-up questions to zero in on what they're looking for.

Me: Much better, thanks. Now, to delve a little deeper, can you give me an example of a first class prompt/query that would get the best output concerning AI and its impact on the economy. I'm thinking from a macro economic context, want the output in a table and want you to give me step by step reasoning. Can you encapsulate that, plus anythnig else that would improve your answer, in an example prompt?

ChatGPT: Certainly. To solicit the most meaningful and well-structured answer from me regarding AI's impact on the economy in a macroeconomic context, while also requesting the information in a table format and including step-by-step reasoning, the query could be phrased as follows:

"Can you provide an analysis of how AI is impacting various sectors of the economy from a macroeconomic perspective? I'm interested in understanding both positive and negative effects. Please present the information in a table format that includes the sector, the positive impact, and the negative impact. Additionally, can you walk me through your reasoning for each point?"

This query includes several elements that can facilitate a high-quality output:

1. Topic Specificity: It asks for an analysis of AI's impact on the economy, focusing specifically on a macroeconomic context.

2. Context Clarification: It specifies that both positive and negative effects are of interest, thus ensuring a balanced viewpoint.

3. Format Specification: It requests that the information be presented in a table, which helps in organizing the data more effectively.

4. Reasoning: It explicitly asks for step-by-step reasoning for each point, ensuring a depth of analysis.

Hopefully, you can see from this that there is an optimised method for getting the most out of ChatGPT. Please feel free to copy and paste the above ChatGPT generated prompt and see what output you get.

That’s it for another week. Thanks for reading and for being a subscriber. Next week is all change! Following feedback from many of you, the format of the newsletter will change completely. I will be doing one newsletter on Wednesday which will be short, snappy news led with some example prompts that you can end to end read in less than five minutes. I’ll then do a deep dive in a second newsletter on a topic of the day which be a single item similar to the Just One Thing section that has proven popular in this version of the newsletter. Thanks again for all the feedback and your continued support. I really appreciate you all.