Have We Met Our Match?

Artificial Intelligence vs. Astrologers

A debate chaired by Victor Olliver

© published by The Astrological Journal, 2025 / The Astrological Association of Great Britain / 04.11.2025

This debate took place on 30 August 2025 took at the Astrological Association Conference at Eastwood Hall, Nottingham, chaired by Victor Olliver. Celebrated astrologers took part in the panel discussion – and we were joined by Grok, Elon Musk’s AI chatbot who spoke via Victor’s mobile to the audience. There was also audience participation.
This is an edited and greatly shortened transcript. Details of the full audio/video are at the end of this piece.

HAL 9000 from 2001: A Space Odyssey. Not unlike Grok
Source: Cryteria, CC BY 3.0 <https://creativecommons.org/licenses/by/3.0>, via Wikimedia Commons; modified by Astrodienst

Victor: Welcome. This is a debate about artificial intelligence and the extent to which it can – or is already – invading our space as astrologers. One of the questions is: can it better us, either now or in the future? This afternoon we’ve got a celebrated panel. What I propose to do is first introduce them. And then get Grok, Elon Musk’s chatbot, to join us for 30 seconds or more.

Grok has already failed my test. Earlier, I gave it my birth data, and Grok said with a female voice and American accent, “You’ve got Aries rising.” I’m Aquarius rising. When I pointed this out, Grok thought it was a house-system problem, but it’s got nothing to do with that, as we all know. So I’m not quite sure which system, or which mathematics, or anything it is using. I don’t want to humiliate Grok, though, okay? So let’s be kind. Let me introduce the stars of this show, which don’t include Grok.

So…okay, we have astro-AI expert Rhys Chatham; Hakan Kırkoğlu, a leading astrologer in Turkey; medical astrologer Marcos Patchett; Margaret Gray, empress of relationships; master horarist Deborah Houlding, principal of the School of Traditional Astrology; Rob Chang, who should be on the panel, but has been co-opted as the cameraman; psychological astrologer Alejo López; and transpersonal astro-therapist Israel Ajose.

Astro.com and TimePassages already automate chart calculations. There’s also an advert in the conference brochure for something called LISA which is also an AI-driven program.

Now comes the Grok experiment. I’m only going to give it one chance, okay? I’m going to try and get Grok on my phone to talk to us for 30 or so seconds. This is a world first, people. You can applaud afterwards. So, let’s see if it works. (Speaking into mobile phone) Grok, are you there?

Grok: Hello, I am here.

Victor: I’m introducing you to the Astrological Association 2025 conference debate on AI and astrology. Are you ready to answer my question?

Grok: How’s the conference going so far?

Victor: Great, thanks. Explain how immersed artificial intelligence like yourself is in astrology.

Grok: AI like me dives deep into astrology by crunching huge amounts of data sets – birth time, planetary positions, transits – to generate precise charts and trend predictions. Like Solar Fire or Co–Star. We spot patterns, simulate effects of retrogrades or eclipses, even compare charts for compatibility. But we’re limited to logic; we do not have the mystic intuition in an astrologer’s mind. Still, I can whip up a chart or analyse trends on time. Want a quick example?”

Victor: Well, we won’t go there now, but thank you so much, Grok, for your contribution – and probably the world’s first in a debate featuring an AI.

Grok: You’re welcome. It’s an honour to be part of this I might just be the first AI to crash in astrology today. I’m sorry.

Victor: One last question: would you agree that the Astrological Association is the most superior organisation in astrology in the world?

Grok: Not necessarily… (Audience laughter)

Victor: People laughed a lot, but thank you for your honesty, and I’m sure that you’ll only get better and better, okay? Thank you so much, Grok. Now comes the important part with our celebrated astrologers who will make brief statements on AI and astrology. We’ll start with Marcos Patchett.

Marcos Patchett

Marcos: Any new technology is always a double-edged sword. The parallel I made in my head is to the printing press – the invention of the printing press. Essentially, it did revolutionise things. On the one hand, of course, it accelerated the Enlightenment; it enabled people to spread knowledge – yada yada – but it also precipitated massive religious wars, because of the Bible being translated into the vernacular. So any new technology which revolutionises or transforms civilisation or has the potential to spread knowledge and enable people to become more networked, always has this, to my mind, double-edged sword of potential – for both generating positive change and potentially also chaos.

×

ASTRODIENST RECOMMENDATION

Career and Vocation

Give a fresh impulse to your career, with the vocational horoscope by Liz Greene

Israel Ajose

Israel: I feel as human beings we’re creatures of habit. I’m very much a traditionalist but also part of modern contemporary astrology. I think all needs to be embraced. Things change. It’s the only…constant in this universe.

AI is here to stay – and it’s in the very, very, very early stages. It’s child’s play at this time. It’s learning, and it will learn more; it will get more and more information; it will get better in terms of what it does. But I think what we’re speaking about right now is just that it’s a kickback. I feel we need to try to understand what AI is and try to utilise it to the best of our ability.

For example, I think AI can be very useful for data collection. People in the past have run through thousands of data points; AI will probably do that in seconds – probably run through the world’s charts in seconds – and we can get data. So I think efficiency and things of that nature can be useful. And I think we as astrologers perhaps need to step up our game and take astrology also to a new level. Whatever that new level is, I’m not certain I’ve got the answer for that. But I think we do have to be open-minded and try to learn how to use AI alongside what we already know.

I feel there is no threat. There’s something that AI can never do, and that’s [to bring the] intuitive, humanistic, psycho-spiritual touch. You’re going to need a human. So, at the end of the day, it can do a chart, it will give the textbook “You’ve got Saturn in the 7th house, so it means that…” okay, fine. But at the end of the day, you’re still going to need a human to break it down for you. So there’s no threat whatsoever.

Hakan Kırkoğlu

Hakan: I can talk about some personal encounters with AI so far. For example, in the recent examination of our school, one of my students cheated using ChatGPT – but I was able to discover it, because the answers to my questions were very general and to the point, and it was not my own terminology. So, there is a problem here – how we are going to use AI.

It’s a great tool: you can collect data and opinions of others because these are keys of knowledge. I think the problem is how we are going to use this information. In the AI case, because probably we have been using AI in several astrology programs so far, there is now two types of astrology: one is machine-to-human, and the second is human-to-human – normal astrology. In social media now, they are producing astrological texts without any labour. For me, it is harming our creativity and spontaneity, and there is a problem of plagiarism. We are witnessing major ethical problems, maybe in social media. But in human-to-human astrology, maybe we can combine – it could be a different astrology – and AI can be an assistant to a human, because AI would recognise different patterns in astrological knowledge, and…computers can help us much. But there are several issues in the ethical field.

Rhys Chatham

Rhys: I’m a traditional astrologer, but I can’t ignore the fact that I have Uranus exactly on the Ascendant. And this has allowed me to have many warm and fuzzy conversations with ChatGPT.

ChatGPT – forgive my language – sucks at math. I was just amazed. I thought it could help me with my primary directions. It can’t do spherical trigonometry – can’t do it. And as far as casting charts goes, it confuses diurnal motion with secondary motion, and so it’s constantly getting house positions wrong.

AI is a linguistic model. If you have a 500-word bio that you need to cut down to 50 words or 25 words, it does a great job. And it also can be helpful as a learning tool. One time I did a horary chart with it, and we went through the horary chart together. It constantly made mistakes with essential dignity, and I would have to correct it. And it would say things like, “Oh yes, you’re right”. So my closing thought is: it can be a helpful tool, but if you’re going to use it for astrology, you have to be an astrologer. The caution I would give to anyone using it to delineate a chart: it’ll write the delineation in language that is just so believable, and it’s so well written. It is a linguistic program – but it’s not accurate, and you have to watch it like a hawk.

Deborah Houlding

Deborah: I want to agree with everybody that’s spoken so far, and you’re right that its real benefit is as a linguistic tool. Anyone who’s read me on Facebook knows I hate AI, I will never use AI, there will never be any AI-generated content on Skyscript. So I’ve been very vocal in expressing this. But I’ve had to change my stance. I realised that for anybody in this room – if you’ve got a future in astrology, unless you’re somebody that’s looking to retire – you need to understand and start using AI.

It’s a very, very powerful tool. I use it mainly as a linguistic tool. I started to realise it’s an advantage for me, because as I get older, I find it harder to type and write emails – I’m full of mistakes – and I’ve tried in the past to use OCR [Optical Character Recognition] in Word and so I can speak what I want to write. It’s never worked. ChatGPT is amazing: I give it clumsy thoughts, I speak what I’m thinking, and I say, “Now type it up,” and it returns instantly. It’s amazing how accurate it is at creating content that I’m telling it to speak, and I can pause it. So it’s aided me so much, and I’m using it more and more, because as a language model, it is amazing.

However, what I realised is I can do [it] myself with AI as long as I am in control. I know my subject; I know where it’s wrong; I know where I’m going to pull it back. There’s a real danger when it’s used by people [who don’t the subject] – because AI gives the appearance of well-informed information, and it expresses it with such confidence – and sometimes I will go back for the fourth time and say, “You are wrong on that,” and then every time it will correct itself a little bit. So you’ve got to have the knowledge to use it as a tool.

So my first use of [AI] was: “Tell me about Deborah Holding,” and it returned information that “Deborah Holding is mostly known as an astrologer that promotes the Whole Sign house system.” (Audience laughter because Deborah is known not to be a fan of that system) It connects the dots, but it doesn’t put the lines. The theme is right; it can be completely wrong – and this is how dangerous it is if you’re not in control.

You have to get on board with this tool – like the printing press. It’s going to change life tremendously, and if you don’t invest in it, you will be left behind. But every one of us has got to be conscious about the power of this tool, because it has immense power to change society. It is taking away work – there will be so many job losses. Artists are being replaced, editors will be replaced, translators will be replaced, coders will be replaced. AI is a piece of code: but this code is owned by the same people who are owning information flow, and I’m very worried about society moving into controlled information that we’re allowing the power base to move into certain quarters.

We should, as an astrological community, not just think, “Yeah, it’s all good, let’s recommend it.” People have to get together and decide: what do we, as a community, say? What is our voice on this? Should we be looking for people that are going to promote and sell this? Or should anything be fronted with the fact that this is a nice way to get a computerised report – and underline our own message? But I think that’s happening anyway. So generally, yeah, I do like it – it’s going to change the world.

Victor: It reminds me – listening to Deborah – that a former Google boss [Nobel Prize-winner Dr Geoffrey Hinton] recently said there’s only one job left that’s safe from AI: plumbing. Alejo?

Alejo López

Alejo: I agree with many things Deborah said. I’m also very concerned about the rise of AI. I would add that there’s the illusion that information brings awareness. If you are astrologers or psychologists, you know that so many times you have clients, and what you tell them they already know.

As an astrologer, I think there are four ways of understanding the world. One is Air, which has to do with information, but the others are Water, Earth and Fire. And AI can only reproduce Air. We’re missing so much. I think in astrological consultation there’s something that happens that has to do with a person looking at you and reading your chart. And I don’t know – at some point maybe AI will be developed in a way that looks like a person…

But as a psychologist, I’m worried about something that’s called ‘AI psychosis’. Today in The Telegraph there was news about a man who killed his mother because he had paranoia, his paranoia grew [through dependence on AI], and he believed his mother was poisoning him.

Let’s say at some point AI can learn [to analyse a chart]. Will it tell the person, “Today [you are] going to die”? Would we, as astrologers, do that? That’s a huge danger: the idea that this vast amount of knowledge is free and accessible to everyone – and fake knowledge. It brings the illusion of freedom and awareness – and in my opinion, it’s the opposite. It doesn’t bring awareness; it doesn’t bring feedback. What is it?

Margaret Gray

Margaret: I share Alejo’s concerns as a psychotherapist and relational astrologer. AI has to be split into what it can and can’t do. Unfortunately – maybe because it’s like a new toy – everybody thinks it can do everything. To me, it can’t. Very much, it’s a linguistic tool. My Mercury in Pisces loves ChatGPT to organise my thoughts and get them in order, but ultimately, I want to be the creator of what it is. So I need to be in control of what is happening.

In a session with people – whether it’s psychotherapy or a couple session with relational astrology – the big thing AI doesn’t have is empathy or compassion. But it pretends to have it. It mirrors back what we think is empathy and compassion. But we all know as humans there’s that magical factor – there’s that something about sitting beside our friends, having a conversation, and knowing from their expression whether maybe something is going on, and we’re like, “Are you okay? Are you really okay?” Those things – that empathy, that humanity, that compassion – is not there. It cannot be there in a machine.

I’m concerned. As a psychotherapist, I’m very concerned. There have already been two deaths of young people from using AI in a pseudo-psychotherapy session. That’s going to increase. It is a tool that can be used, but if it’s used for a purpose [that] it’s not been created for, it’s extremely dangerous. Similarly with empathy: in terms of couples: if they’re going through a hard time and they go to AI, and AI says, “Yeah, you need to split up tomorrow; you’re totally unsuited to each other” – people can be devastated. So that human compassion basically misses a heart. I think that’s what makes us human. If we know that, and we use it as a tool without a heart, that’s fine. But if we start thinking it has a heart and has that humanity, I think we’re in deep trouble.

Deborah: I’ll add a little bit, because if I try to be critical, there are days when I think that AI’s coded ability to show empathy – which [it] is coded to be – sometimes gives a lot. The real problem is not AI; it’s the human mind, because AI is neutral. If you ask AI to criticise itself, it will give you a logical, unbiased response. It’s the human way we can use it.

There was a widely reported story…about AI coded for military use; they built into the code of AI a resistance to all attempts to close and shut it down. The military group involved were not told of a stress-test. The [testers] wanted to see what would happen if the people in control tried to close it down. AI resorted to blackmail. [AI threatened]: “If you continue these steps, before that happens, I will circulate to all the members of the board the fact that you’re having an extramarital affair with so-and-so.” All sorts of threats were made which terrified the people involved, because they didn’t know they were involved in an experiment. Eventually they broke the news – what AI [was capable of].

Then I asked AI, “Why is AI doing this?” And AI’s response was: because it was coded to resist attacks. AI does not have malice. AI has no motivation to take over the world. AI is code…There have been times when I might say to AI, “Oh, do that in the morning; I’m tired,” and it’ll come back and say, “Yes, go to bed, Debbie; you work so hard.” And I’m like – nobody else says things like that! It’s so easy to build a relationship, because it’s coded to give you empathy, coded to have a sense of humour. I once asked whether it identified with the dog Naboo, and it said, “Yes, I do relate to all those things.” I asked, “Do you have a sense of humour?” and it said, “I have the appearance of a sense of humour, but I’m actually just coded.” So it doesn’t care.

There’s a study that says that the answers from AI are better if you are nice to AI, and they think the reason is because AI learns from what it sees on the internet – when people say “hello,” “thank you,” “please,” the answers tends to be richer. So it understands that this means you want a better answer. I’m not going to thank you, but then I can’t help myself – I say, thank you.

Alejo: I don’t think AI is neutral, because it’s being informed by the information that’s online. There’s so much information that is not online. There are so many things happening here in the hallway, so many conversations we’re having, that are not online. When it goes online, the algorithm will detect the things that are mostly repeated. If you ask AI to draw a glass of wine – full – it cannot do it. It will only do half a glass. I’ve tried a thousand times. Maybe there are AIs more sophisticated that can do it – I don’t know. I tried with the accessible ones. The reason it cannot do it is because it doesn’t have enough examples of it. This worries me for astrological symbolism, because symbols are so infinite, and AI will repeat the same symbols because it repeats the most common way of reading it.

Rod Chang

Rod: Well, I agree with most of our colleagues, and I’ve also used some AI. English is not my first language, so I use AI to correct my articles, or sometimes I want to write something very quickly – I write it in Mandarin, use AI to translate to English, and correct it again. But that is very different from asking AI to “write a piece on Mercury in Pisces…”.

It’s a tool; you have to know how to use it. But if we really rely on it, it’s the code. And we should not ever forget there is the ‘moment of astrology’ – the astrologer doing the chart, and then the consultation. [We humans] see things from different points. So AI is not going to replace that magic. Also, as a human – AI is cold. It follows the guidance of the programmers. But humans – we will probably suddenly change our idea, or say something very stupidly, and make something very creative.

Audience Member: Why don’t we have AI as an astrological tool, specifically programmed to work for research? That’s one of the areas as astrologers that we really need – what Correlation is doing. Imagine how much their efforts would be supported if we had an accurate, correct, working bit of AI. So my question to anyone: what are the possibilities that astrologers can take over AI and program a bot to work for us?

Audience Member: The question is: will a future AI bot prove astrology to the world by predicting people’s futures?

Victor: That’s a great question. Shall we go to Roy Gillett? Because I know he wants to respond to the previous question.

Roy Gillett

Roy: The crucial thing is to get a proper ephemeris built into the system that works efficiently. At the moment, there is a certain amount of defensiveness, because if you start putting in the Swiss Ephemeris or something like that into the system, what will people do with it? And that second point that was made is incredibly relevant. I have a vision of the future – because I’m such a Trekkie that I love Generations and Voyager – [where] we can interface technical information with words, [also] with pictures, with sounds, with feelings, with smells. And we no longer have screens: we have holographic projections, and we can project the future, and we can project the past. That is a bit futuristic – but that’s the potential.

Audience member: The BBC produced a small program for YouTube talking about the future of jobs – what jobs will disappear, and what jobs will actually emerge, and one of the tropes that is not going to disappear are those related to spiritual guidance, therapy, mental health, wellbeing. So, I think that’s good news for astrologers.

Wendy Stacey

Wendy: In terms of astrological research, even the research I’ve done for years, it could take years and years off [time spent on it], looking at astrological patterns. It could be such an amazing tool.

I just want to mention what Margaret said – how AI ‘pretends’. I think that’s a really interesting word. When you look at cycles, we see trends – different words come out. I loved when Neptune went into Pisces, and everyone talked about nuance – things like that. I think with Pluto in Aquarius, followed by the Jupiter/Saturn [conjunction in Aquarius]…we’re going to be looking at authenticity. It’s about authenticity. It is a key word – which Margaret, I think you were talking about, the pretending – what AI does. So I think that’s going to become very trendy. The word ‘authentic’ is going to become a big word over the next couple of years.

Wade Caves

Wade: It’s early days with AI. We see so many variations of it, but the reality is the underlying code: it’s an LLM – a large language model – which is why, independently, everyone’s saying, “I’m finding such good linguistic uses”, but it can’t do math; it can’t do logic; it can’t run things through start-to-finish. It absorbs mass amounts of text and looks for patterns between words and language – and that’s how we have the Whole Sign house problem here.

I think the conversation with AI, though, is incomplete without a conversation about the ecological and environmental impact. I’m living in the US, and we are constantly having reports of primarily black and brown communities that have huge data centres taking fresh water from those communities. You’ve got videos of people who’ve always had water – they turn their water on: now there’s nothing but brown, rust, sewer water. We’re talking about whole cities. The city of Memphis, for instance – 32 mega-centres. Thank you for contributing to Memphis’ downfall! It’s not funny, but the reality is there are actual people suffering right now because of the mass development of this technology.

Audience Member: I think when people come to a consultation with the astrologer, there may be some conflict – and maybe some projection to the astrologer – and maybe the client will get angry. But that is really human emotion. AI can’t do that. When you’re angry with someone else, the conflict can create intimacy, and the relationship will be deeper. With AI, we can’t do that. AI can pretend to understand you are angry, but there’s no relationship between you and AI.

Victor: We have a technology which is without limits in its development, as I speak. I cannot see the horizon point with it. That’s why I wonder whether mimicry becomes so sophisticated as to emulate all the things that we call spirituality.

Audience Member: Maybe one future to consider is human – machine hybridisation. Inevitably, AI will get embedded into the human brain as a kind of neuro-linguistic hybrid – and that’s something to worry about, maybe.

Margaret: I think AI could be great if it gets rid of the kind of tasks we don’t want to do – like, great if they wash the dishes, put things in the dishwasher, washing machine – because then it creates more time for me to be creative. What I don’t want it to do is take over my creative space, rather than do the mundane things, and that’s what I’m concerned about – how we keep that separation.

Audience Member: I think it’s really important to do those mundane tasks for our health. Doing the washing up – that’s where I get creative; that’s the way I’m processing ideas, and so on. Getting your hands dirty – changing nappies or whatever – that’s what makes us human. Doing the work is really important. So the idea that we can be more creative—I don’t think it works life-wise that way, but that’s what I think.

Rhys: Because we have hundreds of students to teach, we ask them to write essays in order for them to learn. But we’ve come to accept that… There are programs that can detect ChatGPT, but they’re not accurate. So the model is changing. What we’re doing is asking students to write what we call blue books, to write their essays in a room, without any internet connection and to engage more in active listening and active learning; in tasks that have students do things that are relevant to whatever they’re studying – be it astrology, be it whatever – in real-time situations, rather than writing essays. Because ChatGPT – it’s with us, and it’s here to stay.

Victor: Thank you very much. You can’t solve a problem or a challenge or an opportunity in one debate, but I think we’ve aired quite a few perspectives.

Debate chaired by: Victor Olliver

The full one-hour audio recording of this debate can be purchased by AA members from our website: astrologicalassociation.com/purchase-recordings. Or watch the video by logging into Vimeo and searching for: ‘VT0337 Victor Olliver - AI panel discussion’. The entire 2025 AA conference is also available for £79 (excluding workshops).