- Brain Scriblr
- Posts
- AI for Good, Search for Life Beyond Our World
News
The Chinese firm on Thursday launched QwQ-32B, a smaller ‘reasoning’ model resulting from Alibaba’s exploration of the reinforcement learning techniques that made DeepSeek’s R1 such a powerhouse. More is offered below.
Adobe has unveiled its new AI video generation offerings, allowing subscribers to create five-second videos using text prompts. The service is available through Adobe's Firefly application, which integrates with Creative Cloud and offers plans priced at $9.99 and $29.99 per month.
Alibaba has unveiled its latest AI model, claiming it outperforms DeepSeek-V3, one of the most advanced reasoning models on the market. The new model represents Alibaba’s ongoing push to lead China’s AI race against U.S. tech giants like OpenAI and Google DeepMind.
Meta CEO Mark Zuckerberg announced plans to invest up to $65 billion in artificial intelligence throughout 2025. A significant portion of this investment is allocated for the completion of a major AI data center in Louisiana, supporting Meta’s AI initiatives, including the development of the Llama large language model.
OpenAI launched "Operator," a new AI assistant capable of handling various online tasks, such as ordering groceries and processing ticket purchases. Initially available to ChatGPT Pro users, the Operator functions as a semi-autonomous agent, requiring user input for specific actions like account logins.
SoftBank is reportedly in discussions to lead a $500 million funding round for an undisclosed AI startup. The investment reflects SoftBank’s continued interest in artificial intelligence and its past backing of companies like OpenAI and Anthropic.
Leading AI companies have released autonomous agents that can plan and execute complex tasks in digital environments with limited human involvement. OpenAI’s Operator, Google’s Project Mariner, and Anthropic’s Computer Use Model all perform similar functions, aiming to create AI agents that can undertake a broad range of personal and professional activities.
Nvidia reported a surge in fourth-quarter profit and sales as demand for its specialized Blackwell chips that power artificial intelligence systems continued to grow.
Hong Kong will cut thousands of civil service jobs and boost spending in artificial intelligence as it seeks to tackle an increasing deficit.
AI For Good - ET is Out There
It is perhaps unsurprising that, in the search for intelligent, extraterrestrial life, scientists have turned to machine learning. Here’s what’s going on: Project Galaxia, launched by an anonymous scientist and now supported by a coalition of scientific institutions, aims to leverage both machine learning and quantum cognition to analyze vast amounts of astronomical data, interpret complex radio signals, and uncover potential signs of life beyond Earth.
The idea is that, somewhere in that vast trove of deep space data, machine learning models tuned and trained to detect anomalies and analyze patterns might discover “potential biosignatures, technological signals, and other signs” that point to life beyond Earth. These models could identify patterns in the data that are indicative of life, such as specific chemical signatures or unusual energy emissions.
The project’s findings are currently undergoing a “rigorous” peer review process from a team of scientists around the world. This ensures that the data and conclusions are robust and scientifically sound.
According to the project, initial findings have “baffled” the project’s creators. The complexity and volume of the data have presented unique challenges, but the team is undeterred. This year, Project Galaxia aims to release its first comprehensive analysis of all this data, making 2025 a big year in the hunt for alien life.
Image
A highly detailed astronaut in a realistic, well-textured space suit standing at the center of a vast field of pink cosmos flowers on Mars. The astronaut's suit appears natural, with visible fabric folds, reflections on the visor, and subtle signs of wear. The cosmos flowers are lush and tall, reaching up to the astronaut’s chest, swaying gently in the Martian breeze. The scene is captured from a high aerial perspective, showing the astronaut immersed in the sea of pink flowers against the reddish Martian soil. The sky is clear and bright, with soft sunlight streaming down, casting warm and diffused shadows, creating a dreamy yet realistic atmosphere.
Tools I Use |
|---|
Cudo Compute is a cloud-based service provider that offers high-performance computing, AI, and deep learning solutions. Dubsado is great for contract writing and project management. Folk is the number one AI powered CRM tool. N8N is the most powerful automation tool |
AI in Science Research
The rise of AI in scientific research has sparked enthusiasm, with some claiming that large language models (LLMs) can accelerate discovery. A 2023 paper even suggested that LLMs generate more novel research ideas than human experts. But recent findings challenge this assumption, revealing that much of what appears original is actually sophisticated plagiarism.
A study from the Indian Institute of Technology analyzed LLM-generated research documents and found that 36% contained clear, uncredited similarities to existing work (source). Traditional plagiarism detectors failed to catch these instances, as LLMs often rephrase concepts rather than directly copying them. Only 4% of the papers reviewed were deemed genuinely novel. The researchers warn that LLMs tend to follow predictable patterns, making it possible to detect their influence but also raising concerns about their unchecked use in academic writing.
Beyond plagiarism, LLMs pose another risk: the spread of misinformation. The Oxford Internet Institute warns that AI models are prone to fabrications, biases, and factual errors (source). A study on AI detection failures also found that current methods struggle to reliably distinguish AI-generated content, further complicating the issue (source). While LLMs can assist with language translation and summarization, using them as a primary source of scientific knowledge could erode trust in research.
Science depends on original thinking, rigorous validation, and peer review. LLMs, which predict text based on existing data, are inherently limited in their ability to generate true innovation. If widely adopted without oversight, they could flood academia with misleading or low-value content, making it harder to distinguish meaningful discoveries from AI-generated noise. The Oxford Martin Programme on Misinformation, Science, and Media has highlighted how AI-driven misinformation can impact public understanding of science (source).
AI tools have a place in research, but they must be used with caution. The focus should be on enhancing human expertise, not replacing it. Without proper scrutiny, LLMs risk not advancing science but diluting it.
Reasoning AI and The New Hot Trend in Tech
Move over, ChatGPT! There's a new AI buzzword in town: "reasoning" models. Once a rarity, these smart cookies are now popping up everywhere, with tech giants like OpenAI, Anthropic, and even Alibaba jumping on the bandwagon.
Alibaba's Big Move
The Chinese e-commerce giant just dropped QwQ-32B, a compact powerhouse that's turning heads. Despite being much smaller than its competitors, it's punching well above its weight class. The kicker? Alibaba's sharing it on Hugging Face, giving developers a chance to peek under the hood.
What's the Big Deal?
These new models are like the Sherlock Holmes of AI. They break down problems step-by-step, leading to more logical and accurate results. It's like teaching AI to show its work, not just blurt out answers.
Why Developers Should Care
More Bang for Your Buck: Smaller models that pack a big punch mean faster, more efficient AI applications.
Open(ish) Access: While not fully open-source, Alibaba's move gives developers more room to tinker and innovate.
Improved Accuracy: These models could lead to more reliable AI-powered tools and services.
Business Impact
Cost-Effective AI: Smaller, powerful models could mean lower computing costs for AI-driven businesses.
Competitive Edge: Companies adopting these advanced models early could gain a significant advantage in AI-powered products and services.
New Possibilities: From better customer service bots to more accurate predictive analytics, the potential applications are vast.
The Bottom Line
While Alibaba's stock took a slight dip in New York, the buzz around this new AI approach is undeniable. As these "reasoning" models continue to evolve, they're set to reshape how businesses and developers leverage AI technology. Stay tuned – this is just the beginning of a new AI revolution!
Newsletters I like
TSMC Expands U.S. Investment to $100 Billion Amid AI Trade Turbulence
Taiwan Semiconductor Manufacturing Co. (TSMC), the key supplier behind Nvidia’s GPUs, has pledged an additional $100 billion to its U.S. operations, bringing its total American investment to $165 billion.
The Details:
Announced by President Donald Trump, the move follows the recent unveiling of Project Stargate, a $500 billion AI data center initiative led by SoftBank and OpenAI. TSMC’s investment will fund the construction of three new fabrication plants, two advanced packaging facilities, and a major R&D center, all in Arizona. The company expects the expansion to create tens of thousands of jobs in construction and chip manufacturing.
TSMC has already secured $6.6 billion in grants from the Biden-era CHIPS Act, a federal initiative aimed at strengthening domestic semiconductor production.
The Market Reaction:
The announcement comes as Trump reaffirmed 25% tariffs on Canada and Mexico, a move that rattled markets. The Nasdaq dropped 2.6%, the S&P 500 fell 1.5%, and Nvidia tumbled 8% on Monday, extending its 2025 losses to 15%. TSMC’s stock also slipped over 4%.
“Whether the stock market can survive this change remains to be seen,” wrote Chris Rupkey, chief economist at FwdBonds. “One way or another, tariffs will be a shock for the economy.”
Despite TSMC’s expansion, concerns are growing that the AI-driven stock surge is losing steam. February saw early signs of a broader pullback, and the first week of March suggests the trend may be accelerating.
