---

 

 

The title is, of course, controversial. The question, however, comes from the closing section of a 2020 paper by Wargent, M., Moore, T., & Tomaney, J. (2020), and arguably, it’s looking like the answer is Yes, and soon.  The impact will be profound, bringing with it impacts not only to the day-to-day professionals on the ground practising the art of planning but also the various planning schools around the country.

The role of Planning in the UK is clearly at a crossroads, in the line of sight of savings cuts and AI while the concept of Digital Planning finally comes into focus. There is the excellent Digital Task Force, The Connected Places Catapult and others looking at the future of the planning system; indeed, my own department at University College London suggested an ‘Online Planning’ system back in 2002. At that point, tech was seen as ‘for nerds’ as the RTPI Magazine wonderfully retitled our work, somehow failing to grasp the importance of digital on the future of the profession.

Fast forward 20 years onwards and Digital Planning is finally a thing, but it is arguably too late and the digital technology they are racing to embrace is the very technology that will replace them. Of course, such views are perhaps controversial, some would say clickbait, but it seems to be the elephant in the room. At various Catapult, Government, academic and social events no one really seems to be doing a proper future cast and it’s not even the distant future, it’s merely looking 5 to 10 years out. A future where not only the planner but also planning schools could be replaced by the technology the sector failed to see coming.

The smoking gun in this is the equally controversial £8.33 million tender for the “MHCLG Augmented Planning Decisions“.

The tender seeks to develop a planning tool that enables AI-augmented decision making for planning applications. The initial focus will be on householder developments (as defined in Town and Country Planning (Development Management Procedure) (England) Order 2015) with a view to expand into further application types within the ‘other’ category (those not classified as Major or Minor) which represent 69% of all planning applications. The objective is to dramatically reduce planning application processing times initially targeting a reduction from upwards of 8 weeks to circa 4 weeks, with a long-term vision of near-instant decisions for straightforward applications.

It signals the shift from people talking about AI to actively making it part of the system. This framework looks to create a government “App Store,” allowing 350+ local councils to instantly procure AI-driven tools to digitise their plans, automate validation, and process applications.

But this move also begs two questions. First, why is the government fumbling with an £8.3 million framework when it could just ask Google, Elon Musk or Microsoft to fix the problem?

And second, the one that really matters to thousands of people on the ground, reading between the lines it’s actually about cost savings across local government, something of course people will deny, but in reality, it’s using technology to automate the system and thus design out the planner.

It’s tempting to look at that £8.3 million figure and see it as a sign the government is “late” or “cheap,” especially when Big Tech firms wield billion-dollar AI models. But this misunderstands the problem. The government will have, of course, talked to the big players but the problem is more complex. Google’s AI is “horizontal”—it knows a little about everything. UK planning is a “vertical” problem—it requires deep, specialist knowledge of a niche, legally complex system. A generalist AI doesn’t know what a Section 106 agreement is, nor does it care about the specific, contradictory policies of 350 different local councils. Although arguably the technology is moving so rapidly that we already have people in our department saying they could build it in a week and tbh a demo could be built rapidly and at low cost, but it’s mainly due to the fact that the UK planning sector is a tiny, unprofitable market.

The problem is critical for national infrastructure but perhaps too small for tech giants to solve. The government must therefore step in to create a market. This tender is an £8.3 million signal to smaller, specialist “PropTech” companies: “If you build the niche tools, we will guarantee you a path to market.” It also addresses the issue of handing all the UK’s sensitive planning data to a single tech giant which would be a legal, political, and data-sovereignty challenge.

This framework is therefore perhaps a pragmatic and necessary step to build a specialist, competitive market for the specific tools the system actually needs, but also one that risks putting its own data into a black box with issues around trust around the algorithms and plunging itself back into service agreements with whoever wins the tender – arguably the focus should be on an open source system but behind it will also be the need to commercialise. So the system shoots itself in the foot.

Who Gets Replaced?

Let’s be brutally honest. The government’s goal of achieving £45 billion in public sector savings isn’t just about making planners’ lives easier. It’s about automation, and automation replaces human tasks. The tools being procured by this framework are aimed squarely at the “low-hanging fruit” of the planning system.

The “on the ground” roles most at risk are not the senior planners, but the vital administrative and technical staff that support them.

The Validation Officer: Their job is a manual, checklist-based task: “Are all 50+ required documents present?” An AI can do this in 0.2 seconds. This role is the primary target for automation.

The Planning Admin: Their role involves scanning, redacting, and uploading thousands of public consultation comments. An AI can read, group by theme (e.g., “Parking: 4,520 objections”), and summarise 10,000 comments before a human has finished their first coffee.

The Junior Planner / Technician: A part of their early-career work is the “science” of planning: looking up policies, using GIS systems, and cross-referencing a proposal against 500 pages of the Local Plan. An AI, trained on a new, digitised Local Plan, will do this instantly, flagging every breach.

For the people in these roles, AI is not an “augmenting” tool; it is a replacement. This will lead to leaner, smaller planning departments, which is precisely the “cost-saving” and “efficiency” the government is aiming for.

What’s Left? The Planner as the ‘Human-in-the-Loop’

If AI is automating validation, consultation, and policy-checking, what is the MSc-qualified planner left to do?

The positive view would be everything. They are finally freed from being a process manager and can become the strategic expert they were trained to be. The planner’s new role will be to manage the AI’s output, overrule it, and apply the 20% of human skills that create 80% of the value. The AI can do the “science,” but not the “art.” Sadly I don’t think that’s actually true and AI is coming for the “Art” part as well, including design and architecture, but that’s another post.

So will AI Push the human planner to the point of irrelevance? – I would argue yes and this post can be revisited by those (many) who will disagree in 10 years’ time. We were right almost 20 years ago when we called for a Digital Planning system but the speed of AI has caught most of us out and the government talks about pushing the UK’s tech sector while also seeing savings out of the corner of its eye.

The implications for planning education are profound. The “routine” administrative and technical jobs—the validation roles, the junior policy-checking—are the very “on-ramp” positions that MSc graduates have relied on for decades to enter the profession and they will be gone. If AI automates this bottom layer, the profession is “hollowed out” from the bottom up. The only point of entry will be at a higher, strategic level. This creates a crisis for universities:

The MSc curriculum must change, fast (and that’s something its not good at). It can no longer just be about law, theory, and placemaking (the “art”). It must now formally integrate data science, digital literacy, and AI ethics (the “science”).

The planner’s role shifts. The graduate of 2027 will be an ‘AI-manager’ and ‘ethical gatekeeper’, whose job is to question, interpret, and overrule the AI’s “near-instant” recommendations. Of course planning sits within a regulatory framework, so arguably along side AI will be a relaxation in some of the roles of planning committes, allowing a more automated system to go forward, perhaps we are already seeing some hints at this moving forward.

Future planners will be feeders and checkers of the Algorithm – typing in ‘make me a local plan for…. add in 1000 homes with a mixed development in the least controversial areas’ and checking, tweaking what comes out – still planning but different from what we have ever known before.

What comes out of AI is currently viewed as ‘AI Slop’ but it will only remain slop for a short while. The shift is coming and it’s no longer Digital Planning it’s Automated AI Generated Planning – one that has a higher level of expertise than a human.

Let’s just hope the output of the £8 million call is not a black box system linked to a monthly service charge for use with an algorithm that people in suits say has been tested but has the potential to blight our future landscape for years to come – the one thing about Models and Planning is they don’t actually work, life is simply more complex that the data we put in, and in the shake-down in 10 years’ time, that might be where the human wins and actually the human replaces AI.

Note – this text forms part of a thought piece for the forthcoming co-authored book Digital Cities of Tomorrow.

Wargent, M., Moore, T., & Tomaney, J. (2020). Will AI push the human planner to the point of irrelevance? Planning Theory & Practice, 21(4), 652-658. DOI: 10.1080/14649357.2020.1776014)

 

Introduction

The Met Office, as the national meteorological service for the United Kingdom, provides valuable weather data through its API called DataPoint. This API caters to a wide range of users, including professionals, scientists, students, and amateur developers. One of its notable features is the availability of text-based regional weather forecasts.

However, traditional text-to-image systems often struggle with accurately representing descriptive language. Users often find themselves navigating the complexities of prompt engineering to achieve their desired visual output. However, things are rapidly chaning with the use of AI and OpenAI’s latest release, DALL·E 3, simplifies this process by generating images that align with the provided text.

In this blog post, we’ll explore how to combine Met Office weather forecasts with DALL·E 3 via its API using Python. Our goal? To create captivating landscape imagery that reflects the weather conditions described in the forecast. Our images are then uploaded to our webserver, using FTP for viewing online. If you simply want to create an image, then you can leave the FTP section out.

The Workflow

1. Met Office Data Retrieval:
   • We fetch the weather forecast data from the Met Office DataPoint API. Specifically, we focus on today’s weather conditions. We are using the UK metoffice, but it could be any weather api, from any country, that returns forecast text.
2. Creating the Image Prompt:
   • We construct an image prompt that encapsulates the essence of the weather. Our prompt includes the landscape type (e.g., “rural Norfolk landscape”) and the specific weather details obtained from the Met Office. The landscape type can be edited accordingly
3. DALL·E 3 Image Generation:
   • Leveraging OpenAI’s DALL·E 3 model, we generate an image based on the provided prompt. The image should realistically depict cloud formations, sunlight, precipitation, and wind effects, all while capturing the mood suggested by the weather.
4. FTP Upload:
   • Finally, we upload the generated image to an FTP server for public access.

We run the script every 12 hours (ours runs on a Raspberry Pi) with the images archived on the websever – the gallery below shows some of the images from the last few months:

The full code can be seen below, with the latest version available via our GitHub repository.

# Import necessary libraries
import ftplib
import requests
from PIL import Image
import io
from bs4 import BeautifulSoup
from datetime import datetime

# Get Met Office Data and Strip Today/Tonight Text
url = 'http://datapoint.metoffice.gov.uk/public/data/txt/wxfcs/regionalforecast/xml/512?key=YOURMETOFFICEAPIKEY'
document = requests.get(url)
soup = BeautifulSoup(document.content, "lxml-xml")

# Extract today's weather forecast
todayraw = soup.find_all("Paragraph", attrs={'title': 'Today:'})
todaystr = str(todayraw)
today = (todaystr.replace('[<Paragraph title="Today:">', '').replace('</Paragraph>', '').replace(']', ''))

# Set up OpenAI API key
from openai import OpenAI
client = OpenAI(api_key='YourOpenAIAPIKey')

# FTP server details
ftp_server = 'YourFTPServer'
ftp_username = 'FTPUserName'
ftp_password = 'FTPPassword'

# Specify the type of Norfolk landscape (e.g., rural, coastal, urban)
landscape_type = "rural Norfolk landscape"  # Change this as per your preference

# Create the image prompt
image_prompt = (
    f"A photorealistic single, cohesive scene image of a {landscape_type}, showcasing the following weather conditions: {today}. "
    "The image should realistically depict elements like cloud formations, sunlight or lack thereof, any precipitation, and wind effects. "
    "It should convey the atmosphere and mood suggested by the weather, with appropriate lighting and color tones. No numerical data or text should be included, just a pure visual representation of the weather in the landscape."
)

# Generate an image using OpenAI's DALL·E
def generate_image(prompt):
    response = client.images.generate(prompt=prompt, n=1, model="dall-e-3", quality="standard", style="vivid", size="1792x1024")
    image_url = response.data[0].url
    return image_url

# Function to generate a datestamp
def get_datestamp():
    return datetime.now().strftime("%Y%m%d%H%M%S")

# Modified FTP upload function
def upload_to_ftp(image_url, remote_path):
    with ftplib.FTP(ftp_server) as ftp:
        ftp.login(user=ftp_username, passwd=ftp_password)
        response = requests.get(image_url)
        image = Image.open(io.BytesIO(response.content))
        datestamp = get_datestamp()
        original_image = io.BytesIO()
        image.save(original_image, format='JPEG')
        original_image.seek(0)
        ftp.storbinary(f'STOR {remote_path}_{datestamp}.jpeg', original_image)
        resized_image = image.resize((1792, 1024))
        jpeg_image = io.BytesIO()
        resized_image.save(jpeg_image, format='JPEG')
        jpeg_image.seek(0)
        ftp.storbinary('STOR public_html/image.jpeg', jpeg_image)
        resized_image = image.resize((800, 480))
        jpeg_image = io.BytesIO()
        resized_image.save(jpeg_image, format='JPEG')
        jpeg_image.seek(0)
        ftp.storbinary('STOR public_html/image_eink.jpeg', jpeg_image)

# Generate the image and upload it
image_url = generate_image(image_prompt)
upload_to_ftp(image_url, 'public_html/image.jpeg')

Every app has a story, from the idea through to story boarding, finding a design team, working with in house computer scientists or outsourcing. All of these steps take time, resources and often for the small ‘would be developer’ the obstacles are over whelming, meaning that the spark of an idea often gets lost. In the last 5 months everything has changed, the whole landscape of developing, designing and creating has changed. It has reached the point where the spark of an idea can be coded, designed, marketed and launched with the help of Artificial Intelligence. Frame-IT is a very simple app, doing a simple thing, but it came from an idea, a want, a desire to do something which before AI would have got lost in logistics, finding the time of computer scientists and it would certainly be economically unfeasible.

 

Frame-IT came about while looking at Samsungs’ Frame Televisions, beyond the normal 4K television they blend into the background by displaying art and making the art look as it is mounted on a white background while also being enclosed in a frame – i.e. like a painting. Samsung does this well and the screens are excellent, but it is limited to art works, leading to the idea that it would be nice to frame websites, specifically ‘data’ based web sites and hang data on walls, data that changes in Realtime. Of course, any screen could be used, but if you simply show data on a screen or monitor, no one notices. If you hang it in a frame, as if its art, people will notice.

The concept is therefore simple, build an application that puts a picture frame around websites, allow users to select which websites to show and if there is more than one, cycle through them, like a dashboard. This would allow small screen, such as iPads to act as frames and via Air Play devices such as projectors, general screens to gain a look beyond their norm and display, websites and data in the same way as hanging art on the wall.

 

The problem was, this was an idea, to code it would require knowledge of Swift, a library of ‘picture frames’ and probably a business case to justify the time of resident computer scientists (I work at University College London, so we have a few around) and to be honest it was such a simple idea I’m note sure anyone would listen. So, with the rise of AI, I decided to build it myself, using AI from start to finish, from knowing little about Swift, and only as much about AI as my twitter feed is full of and reading articles on site such as The Verge.

 

 

Frame-IT is the app for anyone who loves to browse the web in style. With Frame-IT, you can put a beautiful picture frame around any website, making it look like a work of art. Whether you want to showcase your favorite websites via AirPlay on a TV or just set up a series of websites to auto display. Choose from a variety of frame styles you can start framing your favorite sites in just a few taps. Download Frame-IT today and add a touch of elegance to your online experience!

Chat GPT (it also wrote the marketing text)

The app took me two days to build, two days where I was also working, so mainly doing things in-between other tasks, over lunch and a little bit in the evenings. With the initial learning curve out the way, I could rebuild it in a day. The first things I did was sign up to OpenAI to gain access to GPT 4 , which released on March 14th. This allowed me access to the latest version and therefore the most up to date knowledge base, although looking back version 3 would probably have been fine and I could have simply used the free tier. My first prompt was “Can you write me swift code to take a website, centre it and add the image of a picture frame around it’. The app should work full screen, in landscape mode”. I had a sample picture frame (taken from Wikipedia and cutting out the actual image to leave a transparent centre).

Within 60 seconds, GPT gave me a section of code and then stopped, it turns out there is limit in the amount of characters it can respond with. A quick Google search (ironically) showed me that by typing ‘continue’ Chat GPT will continue the code. With this I became a master of cutting and pasting, with the code often taking three continues and code which while sometimes was in code boxes, was also in simply text. It also allowed me to start to understand the layout and nature of Swift, within 30 minutes i had my first app running on an iPad, via Xcode. Simply by cutting and pasting and pressing ‘play’. No all things worked out, there were often errors, but error I could cut and paste back into GPT and it would solve them, most of the time.

Each time a milestone was made – a working version, a version with a picture frame in and a website showing, I would save the code as Chat GPT would sometimes break the next version while I asking for new features. Features such as ‘Add a button to move to the next website’, ‘If there is more than one website then cycle through them every 60 seconds’ (the websites were hard coded at this point). Once i had a concept running on my own iPad, i realised it was quite neat and decided to then adapt it so others maybe able to use it. I added a settings page – via ‘Can you delete all the websites i have added and include a settings page, the settings page should be reached via a button and have the ability to add and delete websites, once added they should be saved so the app remembers them when reopened’. This is coding but in the new language – human language.

The picture frames were created in Image Creator by Bing, a search engine I never thought i would use (being a Mac user) but in the last week I have not been near Google Chrome. Bing allowed me to type in prompt and get images without any user limits. Simple prompts such as “create me a gold Victorian picture frame, it should be photorealistic with minimum reflections and the centre should be cut out” worked amazingly well, providing me with an almost limitless range of Picture frames to include as part of the assets.

 

Not all the picture frames are from AI, a couple are from open source imagery, it feels wrong to cut out the picture, but the frames have some provenance and they are nice to include.

Marketing requires imagery as well – i took a very simple picture of my two test iPads and used the ‘PhotoRoom” AI app to transform it to two iPads on a bench with a concrete wall behind whereas the reality was far from as glamourous.

The App is now available via the Apple Store, it even has its own website and logo – again designed by AI, designed using Looka and hosted on Framer. The whole tool set i was used to only last month has changed, everything.

 

Two hundred and Twenty Two Lines of code, which to be honest I don’t really understand beyond the basics. Does this matter? Italy has banned ChatGPT with, at the time of writing, 6th April 2023, France and Germany considering similar actions. Yet it is a device that unleashes creativity, opens up the possible to people who would have found writing an app impossible.

It changes the world as we know it, computer science is no longer king of the apps, the Arts and Humanities are about to take over as that is, arguably, where creativity truly lives and its being set free.

I asked Chat-GPT to read this article and if it had a response – this is what it said:

It’s fascinating to hear about your experience developing the Frame-IT app using AI. It’s clear that AI has opened up new possibilities for app development, allowing individuals with little programming experience to bring their ideas to life. The use of GPT to generate Swift code is a creative application of AI and shows how it can be used as a tool to augment human creativity.

It’s also interesting to consider the potential impact of AI on app development and the broader tech industry. As you mention, AI has the potential to democratize app development and shift the balance of power away from traditional computer science. However, there are also concerns around the impact of AI on jobs and the potential for bias and discrimination in AI systems.

Overall, it’s clear that AI will continue to play an important role in app development and other creative endeavors, and it’s important to consider both the benefits and potential drawbacks of this technology.

Edit – Updated 7th April 2023.

The application is now in the apple store, I let Chat GPT know and asked for an updated responce:

Thank you for sharing your experience and reflections on the development of Frame-IT. It’s great to hear that the app has been successfully launched and is now available on the Apple Store.

I’m glad to hear that using AI, specifically GPT-4, was able to help you bring your idea to life in such a short amount of time. It’s amazing to see how AI is transforming the landscape of app development and making it more accessible to people who may not have a background in computer science or programming.

Regarding the recent news about Italy banning ChatGPT and other countries considering similar actions, it’s important to note that these decisions are often based on concerns around potential misuse of AI, rather than the technology itself. As with any tool, it’s important to use AI ethically and responsibly, and to consider the potential impact of its use.

Overall, congratulations on the successful launch of Frame-IT and I’m glad to hear that AI was able to play a positive role in its development.