?? AI News: Generative AI, ChatGPT, Worldcoin, AI Collapse, and More (Saturday, March 9, 2024)

In this edition of AI News, I cover 41 news about some of the latest developments and trends in the field of artificial intelligence, from generative AI to AI security. I also share some interesting articles and podcasts on AI and society, AI and business, and AI and technology.

??

Teachers use ChatGPT to correct work

Some 3rd through 12th grade teachers are using a ChatGPT-powered grading tool called Writable.

This tool, acquired last summer by Houghton Mifflin Harcourt, is designed to streamline the grading process, potentially offering time-saving benefits for teachers. However, relying on AI for remediation can have drawbacks.

??Some teachers are now using ChatGPT to grade papers | Ars Technica

??

Training a 70b language model at home is now possible

Answer.AI has launched its first project: a completely open source system that allows you to efficiently train a large 70b language model on a desktop computer with two or more standard gaming GPUs (RTX 3090 or 4090). This system will help the open source community release better models. Answer.AI made this its first project because it is a fundamental foundation of its main goal: to help make useful AI available to everyone .

??Answer.AI - You can now train a 70b language model at home

??

Scientists warn of the collapse of AI

Sabine Hossenfelder, an influential scientist, discusses in her recent video the possibility of a collapse in AI creativity. She mentions that although we are used to AI-generated art in the form of text, images, audio and even videos, scientists are warning that AI creativity may soon disappear. She questions what this means for the future of AI and whether human creativity will be in demand after all.

User comments vary, with some expressing skepticism towards AI, while others discuss the possibility of an imminent collapse of AI.

??Scientists warn of AI collapse - YouTube

??

Investigators crack security of AI chatbots with ASCII art

Researchers in Washington and Chicago have developed ArtPrompt, a new way to bypass the security measures built into large language models (LLMs).

ArtPrompt is a tool that generates ASCII art prompts, inducing chatbots to respond to queries they have been programmed to reject.

??Researchers jailbreak AI chatbots with ASCII art -- ArtPrompt bypasses safety measures to unlock malicious queries | Tom's Hardware (tomshardware.com)

??

Why most AI benchmarks tell us so little

Today's most commonly used benchmarks for AI models do a poor job of capturing how the average person interacts with the models being tested.

Many benchmarks used for evaluation are old, from when AI systems were mainly used for research and did not have many real users. Furthermore, people use generative AI in many ways - they are very creative.

??Why most AI benchmarks tell us so little | TechCrunch

??

Testing of Anthropic's new chatbot leaves something to be desired

Startup Anthropic, backed by Google, Amazon and a host of venture capitalists and angel investors, has launched a family of models - Claude 3 - that it claims outperforms OpenAI's GPT-4 in a series of benchmarks. However, TechCrunch argues that the results cited by Anthropic, which are from highly technical and academic benchmarks, do not correspond to the average user's experience.

??We tested Anthropic’s new chatbot — and came away a bit disappointed | TechCrunch

??

Can AI solve science?

Stephen Wolfram questions whether Artificial Intelligence (AI) can solve all the scientific questions that still remain. While the answer is a firm and inevitable no, that doesn't mean AI can't significantly help advance science.

??Can AI Solve Science?—Stephen Wolfram Writings

??

OpenAI key executive played crucial role in ousting Sam Altman

Mira Murati, OpenAI's chief technology officer, raised questions about Sam Altman's management to the board of directors last year, before he was briefly removed from the company. Questions remain about what led the board to make such a dramatic decision.

??Key OpenAI Executive Played a Pivotal Role in Sam Altman’s Ouster - The New York Times (nytimes.com)

??

Meet Inflection-2.5, the world's best personal AI

Inflection has launched Inflection-2.5, an updated model that is competitive with all the world's top LLMs such as GPT-4 and Gemini. Inflection-2.5 is available to all Pi users today, on pi.ai , on iOS, Android, or in the new desktop app. This milestone was achieved with incredible efficiency: Inflection-2.5 approaches the performance of GPT-4, but only used 40% of the amount of computation for training.

??Inflection-2.5: meet the world's best personal AI

??

Reddit implements AI-powered harassment filter

Reddit is upping its AI game again, this time with the implementation of an LLM-powered harassment filter for the benefit of its army of volunteer moderators. The harassment filter is an optional community safety setting that allows moderators to automatically filter posts and comments that are likely to be considered harassment.

??Reddit moderators getting an assist from a new LLM tool ? The Register

???

Yann Lecun discusses Meta AI, open source, limits of LLMs, AGI and the future of AI

Yann LeCun, chief AI scientist at Meta, professor at NYU, Turing Award winner and one of the most influential researchers in the history of AI, participated in Lex Fridman's Podcast.

??Yann Lecun: Meta AI, Open Source, Limits of LLMs, AGI & the Future of AI | Lex Fridman Podcast #416 - YouTube

Extra-Reading:

AI and Society

• AI has spoken: the Apple Car would have been adorable
• Teachers Are Embracing ChatGPT-Powered Grading
• Things really aren’t going to plan for Sam Altman’s strange eyeball-scanning orb
• What if AGI is not coming?
• AI chatbots use racist stereotypes even after anti-racism training
• Police feared ‘brummie’ accent bias in new call bot
• There is no such thing as ‘ethical AI’ — don’t buy into corporate campaigns
• Why people are falling in love with AI chatbots
• ‘My AI twin may get me more modelling work’
• How Porn Stars Are Fighting Back Against AI
• Facebook uses AI to recommend videos: Are Reels about to get even more addictive?

AI in Business

• Check out the best free AI art generators
• 7 Tips For Implementing Generative AI In Your Organization
• Corporate Ozempic: Avoid The AI Side Effects
• Media agencies move beyond buzz: Practical AI takes center stage at Media Buying Summit
• Many students want to learn more about AI, but schools just don’t have the right tools
• Zapier acquires Vowel team and launches AI automation service Zapier Central
• AI Tokens Retain Bullish Momentum, While This Popular Layer-1 Token Offers Strong Contention
• Robotic, on-demand tech for seamless EV charging
• Predictive AI Models Will Make Every Business Smart, Says Pecan CEO
• French Cryptography Startup Zama Raises $73M for Blockchain, AI Privacy Apps
• ITV Bosses Predict Post-Strike Commissioning “Blip” & Describe Generative AI As A “Co-Pilot” For Creatives
• AI is helping Samsung sell a lot more washing machines
• Why 2024 is your year to experiment with AI

AI and Technology

• Nvidia’s next gen AI superchips will reportedly consume up to 1,000W
• The power of pace in technology
• L’AI può essere facilmente utilizzata per realizzare immagini elettorali fuorvianti
• Important AI trends you should be aware of in 2024
• Here’s something else AI can do: expose bad infosec to give cyber-crims a toehold in your organization
• Today’s biggest AI security challenges

AI and Society

??

AI has spoken: the Apple Car would have been adorable

After a decade of rumors and speculation, the Apple Car is dead .

Yahoo Entertainment, 2024-03-07

????

Teachers Are Embracing ChatGPT-Powered Grading

Schools are widely adopting a new tool called Writable that uses ChatGPT to help grade student writing assignments.

Slashdot.org, 2024-03-07

???

Things really aren't going to plan for Sam Altman's strange eyeball-scanning orb

Sam Altman's futuristic Worldcoin project has just been banned in Spain.

Editor's note: 
Not the best illustration photo ("Torre de Belém", Lisbon, Portugal [a totally different country]) :)        

Business Insider, 2024-03-07

?

What if AGI is not coming?

Notes From the Desk: No.

Mindprison.cc, 2024-03-07

??

AI chatbots use racist stereotypes even after anti-racism training

Large language models still demonstrate racial prejudice against speakers of African American English, despite the safety guard rails implemented by tech companies such as OpenAI.

New Scientist, 2024-03-07

??

Police feared 'brummie' accent bias in new call bot

West Midlands Police trialled the non-emergency call-answering AI voice assistant for two months.

BBC News, 2024-03-07

??

There is no such thing as ‘ethical AI’ — don’t buy into corporate campaigns

Misplaced assumptions about what the technology is and what it is not, only serve to reinforce the narratives of ‘AI being too complex’ for lawmakers.

The Indian Express, 2024-03-07

??

Why people are falling in love with AI chatbots

Thursday episodes are all about big topics in the news, and this week they are wrapping up our short series on one of the biggest topics of all: generative AI.

Biztoc.com, 2024-03-07

??

'My AI twin may get me more modelling work'

As the fashion industry grapples with the rise of AI, one model has created an AI twin of herself.

BBC News, 2024-03-07

??♀?

How Porn Stars Are Fighting Back Against AI

Daily Beast, 2024-03-07

??

Facebook uses AI to recommend videos: Are Reels about to get even more addictive?

Meta is shaking things up by infusing AI into Facebook's video recommendations.

PhoneArena, 2024-03-07

AI in Business

??

Check out the best free AI art generators

Of all the many, many ways armchair philosophers have come to define art and the act of human expression, AI-generated images don’t have much in relation.

Quartz India, 2024-03-07

??

7 Tips For Implementing Generative AI In Your Organization

Successfully implementing generative AI in your organization requires a thoughtful approach and focusing on viewing AI as a complementary tool.

Forbes, 2024-03-07

??

Corporate Ozempic: Avoid The AI Side Effects

If AI is to enterprises what Ozempic is to those looking to quickly lose a few pounds, what are the side effects?

Forbes, 2024-03-07

??

Media agencies move beyond buzz: Practical AI takes center stage at Media Buying Summit

At Digiday's Media Buying Summit in Nashville, media agencies addressed the challenges and opportunities of generative AI.

Digiday, 2024-03-07

??

Many students want to learn more about AI, but schools just don't have the right tools

Students want to embrace new technologies like AI, VR and AR, but schools across Europe aren’t delivering.

TechRadar, 2024-03-07

????

Zapier acquires Vowel team and launches AI automation service Zapier Central

Zapier is a name that will be familiar to companies that have explored workflow automation, and like so many other tech firms, it is now grabbing a slice of AI pie.

BetaNews, 2024-03-07

??

AI Tokens Retain Bullish Momentum, While This Popular Layer-1 Token Offers Strong Contention

Biztoc.com, 2024-03-07

??

Robotic, on-demand tech for seamless EV charging

The technology is infrastructure-free, scalable, and allows users to control their charging from an app.

Springwise.com, 2024-03-07

??

Predictive AI Models Will Make Every Business Smart, Says Pecan CEO

Transforming raw data into meaningful insights has always empowered organizations to make informed decisions.

Biztoc.com, 2024-03-07

??

French Cryptography Startup Zama Raises $73M for Blockchain, AI Privacy Apps

French cryptography startup Zama has raised $73 million in a Series A funding round led by Multicoin Capital, Protocol Labs and participation from well-known crypto players such as Polkadot co-founder Gavin Wood, and Solana co-founder Anatoly Yakovenko.

Biztoc.com, 2024-03-07

??

ITV Bosses Predict Post-Strike Commissioning “Blip” & Describe Generative AI As A “Co-Pilot” For Creatives

ITV boss Carolyn McCall has detailed the potential for a post-strike “blip” as commissioning conversations are restarted with U.S. buyers, while the head of ITV Studios described generative AI as a “co-pilot for creatives.”

Biztoc.com, 2024-03-07

??

AI is helping Samsung sell a lot more washing machines

Other than mobile devices, Samsung makes a laundry list of products, including home appliances like washing machines, air conditioners, vacuum cleaners, and a lot more.

SamMobile, 2024-03-07

??

Why 2024 is your year to experiment with AI

QuickBooks Canada has integrated AI into its processes, achieving great success in its marketing strategy.

Thinkwithgoogle.com, 2024-03-07

AI and Technology

??

Nvidia's next gen AI superchips will reportedly consume up to 1,000W

Upgrade the power grid while you're at it.

PC Gamer, 2024-03-07

??

The power of pace in technology

The CEO of Hivemind Technologies, Erik Schmiegelow, explains what steps matter in speeding software development for transformations.

Mckinsey.com, 2024-03-07

??

L'AI può essere facilmente utilizzata per realizzare immagini elettorali fuorvianti

Lo afferma il Center for Countering Digital Hate (CCDH).

Hdblog.it, 2024-03-07

???

Important AI trends you should be aware of in 2024

One of the most exciting developments in AI is the rise of generative AI.

Geeky Gadgets, 2024-03-07

??

Here’s something else AI can do: expose bad infosec to give cyber-crims a toehold in your organization

Singaporean researchers note rising presence of ChatGPT creds in Infostealer malware logs Stolen ChatGPT credentials are a hot commodity on the dark web, according to Singapore-based threat intelligence firm Group-IB.

Theregister.com, 2024-03-07

??

Today’s biggest AI security challenges

98% of companies surveyed view some of their AI models as vital for business success, and 77% have experienced breaches in their AI systems over the past year, according to HiddenLayer.

Help Net Security, 2024-03-07

What was the news that surprised you the most? Please share your thoughts in the comments section below! ?????

For those with technical knowledge and access to the necessary resources (NVidia RTX3050+ GPU | Google Colab is a good alternative (free limited access to GPUs)):

• Want to try some of these solutions locally by yourself? Discover the Open Sources solutions that are making a difference in the world of AI, the ones I have tested, parameterized and do recommend (and some original ad-hoc scripts, too): Filipe MS Bento (fmbento) / Repositories · GitHub

See all editions of this newsletter here: https://www.dhirubhai.net/in/filipebento/recent-activity/newsletter/

Disclaimer: The opinions and views expressed by the news, tools or services presented in this article are those of the authors and do not necessarily reflect the position or opinions of the author of this newsletter. Furthermore, the author has not added tracking codes or affiliate links.

#AI #ArtificialIntelligence #GenerativeAI #ChatGPT #Worldcoin #AICollapse #Writable #AnswerAI #Zama #Inflection #Meta #OpenAI #SamAltman #YannLeCun #LexFridman #Nvidia #Hivemind #GroupIB #HiddenLayer