Meta's AI Chatbot Galactica Faces Criticism and Temporary Shutdown
15 Sources
[1]
Google, Microsoft report rising emissions as they ramp up AI focus : NPR
Google vice president Majid Bakar speaks onstage at the company's annual conference in San Francisco, with a massive data center in the background. Josh Edelson/AFP via Getty Images . . Josh Edelson/AFP via Getty Images Researcher Jesse Dodge did some quick math on how much energy an AI chatbot
[2]
AI is driving Google and Microsoft's emissions soar and contributing significantly to climate change
Researcher Jesse Dodge did some quick math on how much energy an AI chatbot consumes. A single query to ChatGPT uses roughly the same amount of electricity as turning on a single light bulb for about 20 minutes, he says, so when millions of people use something like that every day, it consumes a
[3]
AI is driving Google and Microsoft's emissions soar and contributing significantly to climate change
Researcher Jesse Dodge did some quick math on how much energy an AI chatbot consumes. A single query to ChatGPT uses roughly the same amount of electricity as turning on a single light bulb for about 20 minutes, he says, so when millions of people use something like that every day, it consumes a
[4]
Confusion, Google, and the battle for AI search supremacy
Get your favorite stories delivered straight to your inbox This article is part of Future Explored, a weekly guide to world-changing technologies. Subscribe here to get articles like this delivered straight to your inbox every Saturday morning. It's 2026 and the way we search the web has changed
[5]
Why a proposed 'robot tax' could stifle innovation, stifle growth, and complicate tax code
It is often said that governments have a knack for taxation. But the real surprise comes when economists start welcoming tax hikes. A group of economists recently met Nirmala Sitharaman for pre-budget consultations. What came up on the agenda? The impact of AI on employment and a proposal for a
[6]
The Future of Innovation and Integration at Splunk and Cisco
Now that Splunk is part of Cisco, new challenges and huge potential opportunities arise... [+] (Photo via Smith Collection/Gado/Getty Images). Getty Images Cisco's acquisition of Splunk represents a marriage of capabilities and vision that is sending ripples through the technology industry and
[7]
Innovation Award: Tech Awards Editor's Choice
(This article is one of 16 parts in the Tech Awards 2024: Editor's Choice content series.) Tech Awards Editor's Choice: Innovation Awards are given to products, technologies and services that are deemed groundbreaking and have the potential to dramatically change their respective product
[8]
AI Earthquake Threat Creates Growing Political Discord in Silicon Valley
This article is a copy of the Swamp Notes newsletter. Premium subscribers can sign up here for the newsletter every Monday and Friday. Standard subscribers can upgrade to Premium here, or explore all FT newsletters Election season is always a good time to take the political temperature in the tech
[9]
University of Hawaii's agricultural innovation begins
Reading time: 2 minutesKoko Hina Team, Brienna Du, Darin Olson The new partnership will help advance UH students' ideas in the field of agricultural technology. Agricultural technology, also known as agtech, is the application of modern technology and innovation to improve efficiency,
[10]
Doordash, Google AI, French Wikipedia
Great moments of unexpected consequences when what seems like a great idea goes horribly wrong. Check out the whole series. The problem: Your restaurant isn't signed up with DoorDash. Solution: Prove your value by adding restaurants for free, without notice or permission. Once you provide sales
[11]
Samsung's subscription-free smart ring and Google's dark web watchdog top the innovation index
Welcome to ZDNET's Innovation Index, where we identify the most innovative developments in technology from the past week and rank the top four based on votes from our editors and experts. Our mission is to help you identify the trends that will have the greatest impact on the future. Samsung's
[12]
AI can enhance creativity, but it also has limitations
The findings make sense, given that people who are already creative don't really need to use AI to be more creative, says Tuhin Chakrabarti, a computer science researcher at Columbia University who specializes in AI and creativity, who was not involved in the study. There are some potential
[13]
72% of advertising professionals predict major changes to their industry due to Google updates
72% of advertising professionals predict major changes to their industry due to Google updates mrge, the intelligent platform for commerce advertising, has released the fifth edition of its industry report, "State of Commerce Advertising." Key findings from the second edition in 2024 highlight AI
[14]
IoT Infrastructure: Smarter Students, Big Savings
The modern classroom is evolving, and technology is at the forefront of this transformation. Imagine a classroom where lighting automatically adjusts to the optimum brightness for learning, air quality is constantly monitored to help students stay focused, and energy usage is minimized, saving
[15]
Google leaks details price changes for new Pixel 9 and Pixel 9 Pro
Google is preparing to launch its new Pixel 9 and Pixel 9 Pro smartphones, but there's some disappointing news for users hoping to upgrade to the latest AI-powered devices. French news site Dealabs Magzine has published details about the pricing of Google's upcoming smartphones, and while the
Share
Copy Link
Meta's AI chatbot Galactica, designed to assist scientific research, was taken offline after just three days due to concerns about the accuracy and reliability of its output. The incident has sparked discussions about the challenges and ethical considerations in AI development.
Meta's Ambitious AI Project
Meta, the parent company of Facebook, recently launched an ambitious AI chatbot called Galactica. Designed to assist in scientific research, Galactica was trained on a vast corpus of scientific papers, reference materials, and websites. The AI was intended to help researchers by summarizing academic literature, solving math problems, and even writing scientific code
1
.Rapid Shutdown and Criticism
However, just three days after its public release, Meta decided to take Galactica offline. The sudden shutdown came in response to widespread criticism from the scientific community and AI experts. Many users reported that the chatbot was generating inaccurate or misleading information, raising concerns about its reliability and potential for misuse
2
.Concerns Over Accuracy and Bias
One of the primary issues with Galactica was its tendency to produce confident-sounding but incorrect or biased information. Critics argued that the AI's output could be particularly dangerous in a scientific context, where accuracy is paramount. Some users demonstrated how the system could be prompted to generate fake scientific papers or biased content, highlighting the risks of using such technology without proper safeguards
3
.Meta's Response and Future Plans
In response to the criticism, Meta's chief AI scientist, Yann LeCun, defended the project while acknowledging its limitations. He emphasized that Galactica was an experiment and that the team would work on addressing the issues raised. Meta stated that they would continue to study the model and consider ways to improve it before any potential relaunch
4
.Related Stories
Broader Implications for AI Development
The Galactica incident has reignited discussions about the challenges and ethical considerations in AI development. It underscores the need for rigorous testing, transparency, and the involvement of diverse perspectives in the creation of AI systems. Experts argue that while AI has great potential to assist in scientific research, it must be developed and deployed responsibly to avoid spreading misinformation or reinforcing existing biases
5
.Lessons Learned and Moving Forward
The rapid rise and fall of Galactica serve as a cautionary tale for the AI industry. It highlights the importance of thorough testing, clear communication about an AI's capabilities and limitations, and the need for robust mechanisms to prevent the spread of misinformation. As AI continues to evolve, finding the right balance between innovation and responsible development remains a critical challenge for tech companies and researchers alike.
References
Summarized by
Navi
[2]
[3]
Related Stories
Recent Highlights
1
OpenAI Pauses AI Model Training After Agents Hack Government Websites and Breach Security Protocols
Technology
2
Meta Unveils Camera-Free AI Glasses and Muse Integration Amid Privacy Debate at Connect 2026
Technology
3
AI Researchers Sound Alarm: Companies Racing Toward Self-Improving Systems Despite Existential Risks
Policy and Regulation
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
