- The AI Daily Brief
- Posts
- ChatGPT New Feature Leak?
ChatGPT New Feature Leak?
Plus Google SGE and Bing both get updates
The AI Breakdown First Five - Monday June 12, 2023
Today on the First Five:
5. A ChatGPT Future Feature Leak?
4. Andrew Ng’s Conversations with Bengio and Hinton
3. UK Wants to Lead AI Regulation as US Introduces Two AI Bills
2. Microsoft Bing Chatbot Updates
1. Google Search Generative Experience Expands Access
5. ChatGPT Getting More Ready for Work?
A leaker who had previously shared screenshots of what would become ChatGPT’s recently announced chat sharing feature is back with another leak. This time it’s for something called “workspaces” that also suggests file uploads are coming. Tread carefully, however, as sourcing here is quite limited (to nonexistent).
A Reddit user, known for their past leak of the "chat-sharing" feature, has recently unveiled a screenshot hinting at possible new "workspaces."
These features seem to suggest that users could establish profiles which ChatGPT can conveniently recall, and it also indicates the… twitter.com/i/web/status/1…
— Kenny.AI (@kennyle31)
4:47 AM • Jun 12, 2023
4. Talking AI Risk with a Doomer Skeptic
AI entrepreneur and Coursera co-founder Andrew Ng has been publicly skeptical of AI doomerism. He recently asked for an honest, open conversation about what the risks are, and has subsequently talked with two of the three 2018 Turing Award winners Yoshua Bengio and Geoffrey Hinton. That hasn’t stopped the third Turing Award winner from that year for remaining incensed at the whole discourse:
@AndrewYNg We all agree that we need to arrive at a consensus on a number of questions.
I agree with that LLM have *some* level of understanding and that it is misleading to say they are "just statistics."
However, their understanding of the world is very superficial, in… twitter.com/i/web/status/1…— Yann LeCun (@ylecun)
5:29 PM • Jun 11, 2023
3. Which “United” Country Will Lead AI Regs?
British PM Rishi Sunak recently told London Tech Week that he wanted the UK to be “not just the intellectual home, but the geographic home of AI safety regulation.” Meanwhile in the US, bipartisan groups of senators last week introduced two separate AI bills — interestingly both biting off discreet issues (such as disclosure when the US gov’t is using AI) rather than going for something comprehensive.
With £100 million for our AI Taskforce, we are dedicating more funding to AI safety than any other country in the world.
We’ll also work with @DeepMind, @OpenAI and @Anthropic who will give us early or priority access to models for research and safety purposes.
— Rishi Sunak (@RishiSunak)
11:14 AM • Jun 12, 2023
2. Desktop Voice Control for Microsoft Bing Chatbot
The feature cascade continues as Microsoft recently announced a number of updates for its Bing chatbot, including notably voice control via desktop. Microsoft also published a set of “AI customer commitments” that they say are designed to help their customers implement AI responsibly.
🚨BREAKING: Microsoft’s Bing chatbot is getting a massive upgrade!
The new Bing Chat Features Include -
1️⃣ Enhanced visual
2️⃣ Citations
3️⃣ Graphic-centric answers
4️⃣ More Image creating
5️⃣ Chat history export
6️⃣ Third-party plugins
7️⃣ MultimodalityThis means...
— ⟁Vinnie Moura (@vinniemourax)
8:53 AM • May 4, 2023
1. Google Search Generative Experience Rolls Out Further
Google’s I/O event showed off their new Search Generative Experience or SGE. Two weeks ago they started a waitlist and they now appear to be expanding access, promoting the new features on YouTube and Twitter. They also say that the generative search experience is now twice as fast.
Something Cool
You likely saw this last week, but if not, a Reddit user on the (currently dark) StableDiffusion subreddit posted these working artistic QR codes generated with AI. BTW, the subreddit is dark as part of a protest against changed API fees that were driven in large part by — you guessed it — AI companies.
AI Generated QR Codes!!! - This is probably one of the best Controlnet usecases I've seen so far (Yes they work) (thx @BenTheEgg for the find!) reddit.com/r/StableDiffus…
— Ahmed Dahbi (@dahbiahmed)
5:51 PM • Jun 5, 2023
The Latest AI Breakdown Episodes
Thanks for reading! Let us know how today’s edition hit:
How was today's newsletter? |
The AI Breakdown podcast - https://pod.link/1680633614
The AI Breakdown YouTube - https://www.youtube.com/@theaibreakdown
The AI Breakdown Discord - https://bit.ly/aibreakdown
Signing off from the future’s past - NLW