- AI For All
- Posts
- Google Cloud Next 2023 and ChatGPT Enterprise
Google Cloud Next 2023 and ChatGPT Enterprise
PLUS: OpenAI Legal Battles
AI For All
September 01, 2023
Hello readers,
Welcome to another edition of This Week in the Future where the spotlight is on key developments from Google Cloud Next and OpenAI. From new AI-optimized infrastructures to legal challenges in AI training, this week is filled with topics that are likely to shape the future of the AI landscape.
As always, thanks for being a subscriber! We hope you enjoy this week’s content — for a video breakdown, check out the episode on YouTube.
Let’s get into it!
Google Wants It All
First up is this year’s Google Cloud Next, where the tech giant revealed numerous AI-driven products, notably new TPUs, more powerful A3 VMs, and latency-reducing network upgrades. The key takeaway: Google's infrastructure is providing more computing power at a lower cost.
Vertex AI
Vertex AI saw several minor updates, largely in generative and chat modeling. Google's own GPT variant, PaLM 2, has been updated to support 32K tokens, which seems to be the golden number for a lot of companies. It may be that this is the point at which usability and profitability intersect very nicely.
Digital Watermarking With SynthID
Google DeepMind introduced SynthID, a digital watermarking tool for AI-generated images made with Imagen. This watermark remains detectable even after multiple modifications to the image such as color filters or lossy compression (like JPEG).
It’ll be interesting to see if more companies that offer generative AI products will introduce watermarking like this not just for images but for text and videos as well.
Duet AI: A Copilot for Google Apps
Google is also pushing hard in consumer AI with its Duet AI platform. Integrated into Google Workspace apps, Duet AI can now take meeting notes, auto-translate captions in 18 languages, and summarize documents. This should definitely prove to be a productivity booster for anyone using the Google suite of products for work.
BigQuery Studio
For those using BigQuery for data storage, Google announced BigQuery Studio, a unified workspace that facilitates collaborative machine learning projects. It puts notebooks in the same place so that data scientists can collaborate and keep all of their documentation and work in one place.
Our Take
Based on the announcements from Google Cloud Next, it's evident that Google is plowing ahead with its AI strategy. They are not content with being the third best after AWS and Microsoft Azure.
The tech giant is making aggressive strides both in enterprise solutions and consumer-facing applications. Keep an eye on Google’s next moves, because it’s clear they want nothing more than to dominate the space.
OpenAI's Enterprise Push and Legal Struggles
Next up is OpenAI’s expansion into the enterprise space, where they are working to address much of the feedback around security and compliance to see more company adoption. But also on their plate are some legal battles that could likely alter the future of AI model training.
ChatGPT Enterprise Unveiled
OpenAI introduced ChatGPT Enterprise, which features faster access to GPT-4, a 32K-token limit, and enhanced security features including SOC 2 compliance. This suite of features could make OpenAI tools more attractive for enterprise adoption.
One key piece is that all of the conversations had through ChatGPT Enterprise will not be used for model retraining. This data use is completely private.
Copyright Battles
OpenAI faces lawsuits from authors claiming that ChatGPT produces unauthorized versions of their copyrighted work. OpenAI has countered by saying their use is transformative, teaching models about human language rather than plagiarizing.
The company argues that their tech serves the public interest in tasks like workplace efficiency. OpenAI has cited legal precedents that statistical data used in training aren't protected under copyright law.
The stakes? These cases could set a precedent for how AI companies use large datasets for training, possibly restricting such usage and giving an advantage to unauthorized models.
Our Take
Amidst this, we're forced to ponder: Is machine learning fundamentally different from human cognition? Some might say copyright or IP depends on the notion of ‘free will’ — that you are indeed the author or upstream cause of your thoughts and actions. However, if human cognition is more akin to a complex algorithm that learns via input from external sources, then it will be tricky to sanctify human-generated output over AI-generated output.
We recommend keeping tabs on these lawsuits as OpenAI continues to push disruptive technology. The legal details of the wildly successful deep learning/big data paradigm might see changes in the near future.
🔥 Rapid Fire
IBM and Salesforce team up on trustworthy AI
OpenAI on path to $1 billion in revenue
SoftServe launches AI lab to accelerate enterprise adoption
Pentagon to begin AI trust initiative
US Copyright Office wants your thoughts on AI
IBM Introduces Watsonx Your Business
AI21 Labs raises $155M for enterprise AI
China trying to make its own AI chips amidst sanctions
🎙️ The AI For All Podcast
This week’s episode featured Raghu Ravinutala, the CEO and Co-Founder of Yellow.ai, who discussed how any high-touch business can improve customer satisfaction and boost ROI by adding AI to their customer experience (CX) and what enterprises need to know to start optimizing their business with LLMs!
📖 What We’re Reading
This week’s handpicked insights paint a clear picture of the divided reaction to AI. The technology’s sudden arrival has been met with enthusiasm by some and anxiety by others.
Business Adopting AI Risk a ‘Trust Gap’ with Customers – Salesforce Report (link)
“Notably, consumers have become much less open to using AI over the last year. While 73% of business buyers and 51% of consumers are open to the use of AI to improve their experiences, those figures have dropped significantly since 2022’s survey, from 82% and 65%, respectively.”
Growing public concern about the role of artificial intelligence in daily life — Pew Research (link)
“Overall, 52% of Americans say they feel more concerned than excited about the increased use of artificial intelligence. Just 10% say they are more excited than concerned, while 36% say they feel an equal mix of these emotions.”