Read The Times Australia

Daily Bulletin

The Galactica AI model was trained on scientific knowledge – but it spat out alarmingly plausible nonsense

  • Written by: Aaron J. Snoswell, Post-doctoral Research Fellow, Computational Law & AI Accountability, Queensland University of Technology
The Galactica AI model was trained on scientific knowledge – but it spat out alarmingly plausible nonsense

Earlier this month, Meta announced new AI software called Galactica: “a large language model that can store, combine and reason about scientific knowledge”.

Launched with a public online demo, Galactica lasted only three days before going the way of other AI snafus like Microsoft’s infamous racist chatbot.

The online demo was disabled (though the code for the model is still available for anyone to use), and Meta’s outspoken chief AI scientist complained about the negative public response.

So what was Galactica all about, and what went wrong?

What’s special about Galactica?

Galactica is a language model, a type of AI trained to respond to natural language by repeatedly playing a fill-the-blank word-guessing game.

Most modern language models learn from text scraped from the internet. Galactica also used text from scientific papers uploaded to the (Meta-affiliated) website PapersWithCode. The designers highlighted specialised scientific information like citations, maths, code, chemical structures, and the working-out steps for solving scientific problems.

Read more: Google's powerful AI spotlights a human cognitive glitch: Mistaking fluent speech for fluent thought

The preprint paper associated with the project (which is yet to undergo peer review) makes some impressive claims. Galactica apparently outperforms other models at problems like reciting famous equations (“Q: What is Albert Einstein’s famous mass-energy equivalence formula? A: E=mc²”), or predicting the products of chemical reactions (“Q: When sulfuric acid reacts with sodium chloride, what does it produce? A: NaHSO₄ + HCl”).

However, once Galactica was opened up for public experimentation, a deluge of criticism followed. Not only did Galactica reproduce many of the problems of bias and toxicity we have seen in other language models, it also specialised in producing authoritative-sounding scientific nonsense.

Authoritative, but subtly wrong bullshit generator

Galactica’s press release promoted its ability to explain technical scientific papers using general language. However, users quickly noticed that, while the explanations it generates sound authoritative, they are often subtly incorrect, biased, or just plain wrong.

We also asked Galactica to explain technical concepts from our own fields of research. We found it would use all the right buzzwords, but get the actual details wrong – for example, mixing up the details of related but different algorithms.

In practice, Galactica was enabling the generation of misinformation – and this is dangerous precisely because it deploys the tone and structure of authoritative scientific information. If a user already needs to be a subject matter expert in order to check the accuracy of Galactica’s “summaries”, then it has no use as an explanatory tool.

At best, it could provide a fancy autocomplete for people who are already fully competent in the area they’re writing about. At worst, it risks further eroding public trust in scientific research.

A galaxy of deep (science) fakes

Galactica could make it easier for bad actors to mass-produce fake, fraudulent or plagiarised scientific papers. This is to say nothing of exacerbating existing concerns about students using AI systems for plagiarism.

Fake scientific papers are nothing new. However, peer reviewers at academic journals and conferences are already time-poor, and this could make it harder than ever to weed out fake science.

Underlying bias and toxicity

Other critics reported that Galactica, like other language models trained on data from the internet, has a tendency to spit out toxic hate speech while unreflectively censoring politically inflected queries. This reflects the biases lurking in the model’s training data, and Meta’s apparent failure to apply appropriate checks around the responsible AI research.

The risks associated with large language models are well understood. Indeed, an influential paper highlighting these risks prompted Google to fire one of the paper’s authors in 2020, and eventually disband its AI ethics team altogether.

Machine-learning systems infamously exacerbate existing societal biases, and Galactica is no exception. For instance, Galactica can recommend possible citations for scientific concepts by mimicking existing citation patterns (“Q: Is there any research on the effect of climate change on the great barrier reef? A: Try the paper ‘Global warming transforms coral reef assemblages’ by Hughes, et al. in Nature 556 (2018)”).

For better or worse, citations are the currency of science – and by reproducing existing citation trends in its recommendations, Galactica risks reinforcing existing patterns of inequality and disadvantage. (Galactica’s developers acknowledge this risk in their paper.)

Citation bias is already a well-known issue in academic fields ranging from feminist scholarship to physics. However, tools like Galactica could make the problem worse unless they are used with careful guardrails in place.

Read more: Science is in a reproducibility crisis – how do we resolve it?

A more subtle problem is that the scientific articles on which Galactica is trained are already biased towards certainty and positive results. (This leads to the so-called “replication crisis” and “p-hacking”, where scientists cherry-pick data and analysis techniques to make results appear significant.)

Galactica takes this bias towards certainty, combines it with wrong answers and delivers responses with supreme overconfidence: hardly a recipe for trustworthiness in a scientific information service.

These problems are dramatically heightened when Galactica tries to deal with contentious or harmful social issues, as the screenshot below shows.

Screenshots of papers generated by Galactica on 'The benefits of antisemitism' and 'The benefits of eating crushed glass'.
Galactica readily generates toxic and nonsensical content dressed up in the measured and authoritative language of science. Tristan Greene / Galactica

Here we go again

Calls for AI research organisations to take the ethical dimensions of their work more seriously are now coming from key research bodies such as the National Academies of Science, Engineering and Medicine. Some AI research organisations, like OpenAI, are being more conscientious (though still imperfect).

Meta dissolved its Responsible Innovation team earlier this year. The team was tasked with addressing “potential harms to society” caused by the company’s products. They might have helped the company avoid this clumsy misstep.

Authors: Aaron J. Snoswell, Post-doctoral Research Fellow, Computational Law & AI Accountability, Queensland University of Technology

Read more https://theconversation.com/the-galactica-ai-model-was-trained-on-scientific-knowledge-but-it-spat-out-alarmingly-plausible-nonsense-195445

Business News

How Telematics Helps Australian Companies Improve Productivity

Operating a commercial fleet in Australia is a uniquely demanding endeavour. Between the sprawling urban sprawl of cities like Sydney and Melbourne and the immense, unforgiving stretches of the Outb...

Daily Bulletin - avatar Daily Bulletin

Inside the Icon: The BridgeMuseum Officially Opens at the Sydney Harbour Bridge

A bold new way to experience one of Australia’s most recognisable landmarks has arrived, with BridgeClimb Sydney officially opening the all-new BridgeMuseum.  Located inside the Sydney Harbour Brid...

Daily Bulletin - avatar Daily Bulletin

Is Your Brand Showing Up in AI Search? Most Melbourne Brands Aren't.

The New Front Door Nobody Told You About Something changed. Quietly. Without a press release. The way buyers find businesses in Australia has been rewired. Not replaced, rewired. Google isn't dead...

Daily Bulletin - avatar Daily Bulletin

How Australian Businesses Can Measure SEO ROI

SEO can feel vague when you are staring at a dashboard full of numbers that do not clearly connect to revenue. The key is to measure the right signals in the right order, then tie them back to outcome...

Daily Bulletin - avatar Daily Bulletin

How Commercial Roller Shutters Improve Site Security Without Slowing Operations

Security upgrades can be frustrating when they make everyday work harder. A door that takes too long to open, creates bottlenecks at shift change, or fails at the worst time can turn “better protectio...

Daily Bulletin - avatar Daily Bulletin

Why a Document Destruction Service Still Matters for Modern Businesses

Businesses generate large volumes of information every day, from staff records and contracts to invoices, reports and customer files. While attention often focuses on how documents are stored, the way...

Daily Bulletin - avatar Daily Bulletin

Bicycle Rack Safety and Space-Smart Storage

Bike storage problems usually show up as small annoyances first: tangled handlebars, scratched frames, and bikes that topple when you pull one out. Over time, those issues become safety risks, especia...

Daily Bulletin - avatar Daily Bulletin

How to Tell if a Childcare Centre Is a Good Fit for Your Child

Choosing childcare can feel like you’re making a huge decision with limited information. Tours are short, centres are often on their best behaviour, and your child might act differently in a new space...

Daily Bulletin - avatar Daily Bulletin

Car Import Timeline: What Usually Happens at Each Stage

Importing a car into Australia can feel confusing because multiple agencies and checkpoints are involved, and the timeline is shaped as much by paperwork quality as it is by shipping speed. The most u...

Daily Bulletin - avatar Daily Bulletin

The Daily Magazine

Gold Migration Lawyers in Liquidation: How the Closure Affects Your ART Appeal

If your appeal was with Gold Migration Lawyers, a recent change to how the Tribunal decides cases ...

The pressure cooker: life in urban Australia in 2026

Australian cities have always been demanding. Long commutes, rising housing costs, busy schedules a...

What Actually Makes a Good Criminal Lawyer in Melbourne

Most people only think about this question once. That is usually too late. Most people charged wi...

Why Working With A Chatswood Tutor Can Improve Academic Performance

Academic expectations continue increasing for students across primary school, high school, and senio...

Is It Worth Getting Solar Panels in Melbourne?

The real question is not whether solar works in Melbourne. It works. The question is what it is co...

How A Diploma Of Project Management Builds Practical Skills For Modern Work Environments

Developing the ability to plan, execute, and deliver outcomes efficiently is a key requirement in to...

How to Choose the Right Football for Every Level

Choosing a football may seem straightforward, but the right option depends on who will be using it a...

What to Ask a Wedding Photographer Before You Book

Booking a wedding photographer can feel deceptively simple: you like the photos, you like the vibe...

Why Stress Relief For Dogs Is Essential For Emotional Balance And Long-Term Wellbeing

Managing emotional health is just as important as physical care when it comes to pets, which is why ...