🦸 PwC taps into AI's powers
PLUS: AI funding trends upwards
New here? Welcome to NAB, and make sure to subscribe to stay informed on the latest developments in AI!
Happy Friday, fellow humans 👋
Just wanted to shout out Elon Musk for naming his bot detection program “Not A Bot.”… our website hits have gone through the roof 😅 Appreciate you, Elon!
We’ve also got some interesting AI updates to share with you from the week:
Open-source challengers to OpenAI’s GPT-4V
PwC taps OpenAI’s ChatGPT to assist in the office
Could AI usher in a new age of bioweapons?
And more! Let’s dive into it…
P.S. Have you ever wanted to build a chatbot to help with math problem sets? If yes, check out this awesome guide on turning Vercel's open-source AI chatbot into a personal math tutor!
Have anything you’d like to share with over 40k AI enthusiasts?
🧵 In today's edition:
🗞️ Your Weekly Round-Up
🎙️ BWAI Episode: Jaime Cuffe of Retool
🤑 AI Fundraising News
🛠️ AI Tools to Check Out
🐦 Tweet Post of the Day
🗞️ Your Weekly Round-Up
1️⃣ Nvidia focuses on robotics
Nvidia is embracing generative AI in its robotics offerings to accelerate its adoption among roboticists.
Why? With 1.2 million developers interfacing with Nvidia's AI and Jetson platforms, including AWS, Cisco, and John Deere, the impact is substantial.
To facilitate this, the company has released the Nvidia Isaac ROS 2.0 and Nvidia Isaac Sim 2023 platforms, both of which use generative AI.
They have also launched the Generative AI Playground for Jetson, which provides developers access to open-source large language models.
The idea is to offer a more natural language interface for systems and to help them determine a course of action in unforeseen circumstances, accelerating deployments of AI at the edge.
Read more: Techcrunch
2️⃣ Open source challengers to OpenAI's 'multimodal' GPT-4V
OpenAI's GPT-4V is a 'multimodal' model that can understand both text and images and has been praised for its ability to provide instructions that are easier to show than tell. For example:
It can provide step-by-step instructions on how to fix a bike.
It can also suggest recipes that can be prepared using ingredients from a pictured fridge.
But now, a pair of open-source projects have released similar models that can accomplish many of the same things as GPT-4V.
LLaVA-1.5 is one of the first models that can run on consumer-level hardware. It performed well on specific tasks like object detection and contextualizing images.
Adept's Fuyu-8B focuses on understanding unstructured data and can locate specific elements on a screen. However, it does not come with built-in moderation mechanisms, which could be an issue.
Read more: Yahoo Finance
3️⃣ OpenAI builds an AI image detector
OpenAI is developing a tool to accurately detect if an image was created by their AI service, DALL-E.
The tool is said to be 99% reliable and is being tested before its public release. This might be the real deal!
OpenAI is committed to improving software that detects whether text, audio, or images were made using AI, given the potential for AI-generated content to manipulate or fabricate news reports.
Read more: Yahoo Finance
4️⃣ Square is all in on AI
In the wake of a challenging financial year, Square is looking to revitalize its platform by launching ten new generative AI features.
Among them:
A menu generator.
A website copy generator.
An email copy feature.
Automatic item descriptions for seller catalogs.
While these tools are designed to save time, user feedback remains critical due to the high probability of hallucinations.
Read more: Techcrunch
5️⃣ Nvidia and Foxconn unite to develop software for self-driving cars
Nvidia and Foxconn have announced the opening of AI factories that will build electric vehicles and design software powering fleets of self-driving cars.
The factories will utilize Nvidia's GPU computing infrastructure and will use Nvidia's latest GH200 Grace Hopper Superchip to process and transform large amounts of data in AI models and tokens.
The AI factory will also develop software for cars like the Foxtron Model B, which will send back data to update the self-driving software.
Read more: Gizmodo
6️⃣ PwC taps OpenAI’s ChatGPT to assist in the office
PwC is joining forces with OpenAI to “streamline services, reduce costs, and increase productivity,” making PwC the first of the Big Four audit firms to partner with OpenAI.
The integration of AI will assist in complex tax, legal, and human resources tasks, such as due diligence on companies, compliance identification, and decision-making in business transactions.
PwC also stated that despite recent measures like pay freezes and job cuts due to the professional services industry's economic slowdown, this partnership is not expected to result in job cuts in the near term.
Is there any truth to that statement? I guess only time will tell.
Read more: Bloomberg
7️⃣ AI funding trends upwards
The AI investment landscape is thriving, with the AI sector raising a remarkable $17.9 billion in Q3 2023, up 27% YoY.
This substantial funding, however, illustrates a stark contrast to the overall VC industry, which faced a 31% drop in deals, landing at $73 billion globally during the same period.
The rise in AI funding comes as no surprise, especially with generative AI technologies like ChatGPT gaining prominence, generating photo-realistic images and human-like text based on minimal prompts.
Many VCs see the current AI boom as a pivotal moment, akin to the emergence of the consumer internet.
And while AI is in the spotlight, many tech companies are integrating AI capabilities without needing to redefine their identity. The key is to adapt and keep progressing.
Read more: Bloomberg
8️⃣ Could AI usher in a new age of bioweapons?
The RAND Corporation recently released a report warning that AI could be used to create bioweapons.
Details:
While AI today cannot provide instructions on creating such weapons, it can bridge gaps in knowledge that have prevented their successful creation in the past.
The researchers used LLMs to simulate scenarios in which AI discussed budgeting for bioweapons and identifying potential agents of spread.
The report also suggests that the lack of government oversight over AI could lead to a gap in regulation that terrorist groups could exploit.
Overall, it’s important to note that while AI has transformative potential, it also carries risks that demand vigilant attention and regulatory action.
Read more: Gizmodo
9️⃣ Google India taps gen-AI to help users ‘search’ better
Google has unveiled a set of generative AI-focused features in India to enhance user experiences.
These include:
Access to summaries of over 100 government-led schemes in both Hindi and English, making vital information more accessible.
Provision of localized and visual search generative experiences (SGE) by incorporating images and videos into search results.
The company's expansion of SGE features in India, including language toggles and dedicated slots for generative AI-powered search results, continues to make the search experience more informative and user-friendly.
Read more: Techcrunch
🔟 Microsoft is all about safety
Azure AI Content Safety is a new product from Microsoft that equips businesses with tools to deploy generative AI more safely.
The technology is customizable for different use cases, such as classroom education or gaming, and can help protect users of AI-powered products from inappropriate or harmful content.
Microsoft has nearly 350 people working on responsible AI, and Azure AI Content Safety is one example of how the company is putting its principles for creating AI responsibly into practice.
Read more: Microsoft Blog
🎙️BWAI Episode featuring Jaime Cuffe of Retool
Hey Team! ICYMI, our latest BWAI episode featuring Jamie Cuffe of Retool is out now!
Catch this episode on YouTube, Spotify, or Apple Podcasts today!
🤑 AI Fundraising News
Pantomath raises $14M in Series A funding to build a data pipeline observability and traceability platform.
Reality Defender raises $15M to bolster AI-driven deep-fake detection.
Overstory raises $14M in Series A funding to mitigate wildfires, protect biodiversity, and help people make decisions about natural resources.
Nirvana raises $57M in Series B funding to use telematics data to optimize routing, improve driver coaching, and make the insurance process easier.
Statement raises $12M in Seed funding to allow businesses to gain real-time visibility into their financial data by reducing manual workflows and providing more accurate cash-based data.
Evident Vascular raises $35M in Series A funding to develop an intravascular ultrasound platform designed initially for peripheral vascular interventions, with future capability for coronary procedures.
Hayden AI raises $53M in Series B funding to provide data-driven intelligence applications that help businesses and governments solve real-world problems, such as optimizing traffic operations.
Leucine raises $7M in Series A funding to digitize pharmaceutical manufacturing by leveraging AI technologies to optimize traditional procedures and replace paper-based records with a digital-first approach.
🛠️ AI Tools to Check Out
Impaction AI: Helps product teams analyze conversational data generated from LLM-powered apps. Check them out!
Lakera Guard: The fastest and easiest way to safeguard your LLM-powered applications against prompt injection attacks, hallucinations, data leakage, and toxic language. Check them out!
🐦 Tweet Post of the Day
A great tip for anyone looking to build in AI! 👇️
Not "AI" because it's cool to follow the trend, but rather, because it's remarkably well-integrated into an already great product. Remember, even with this breakthrough tech, it's always about Making Things People Love.
— Alexis Ohanian 🇦🇲 (@alexisohanian)
6:02 PM • Oct 15, 2023
What did you think of today's newsletter? |
And that does it for today's issue.
As always, thanks for reading. Have a great day, and see you next time! ✌️
— Haroon, (definitely) Not A Bot, and @haroonchoudery on Twitter
P.S.: Check us out on our socials - snacks on us if you also hit follow. :)
Autoblocks | Twitter | LinkedIn | YouTube