Bill Franks (@billfranksga) profile picture

Bill Franks

@billfranksga

Analytics & big data focused speaker, blogger, consultant, and author **CURRENTLY INACTIVE** See my LinkedIn page: https://teapawt.net/t.co/vD557shLX7

bill-franks.com
Joined February 2011
  • Tweets9.6K
  • Following111
  • Followers4K
90Media
Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)Photo by Bill Franks (@billfranksga)
Part 3 of 3: If you’re interested, I’ll continue to post content on LinkedIn. My LinkedIn page is here: linkedin.com/in/billfranksga
Bill Franks - Analytics Advisory Partners | LinkedIn

Bill Franks - Analytics Advisory Partners | LinkedIn

Bill Franks is an internationally recognized thought leader, speaker, consultant, and… · Experience: Analytics Advisory Partners · Education: North Carolina State University · Location: Marietta ·...

linkedin.com
2
234
Part 2 of 3: My pause has nothing to do with the recent changes at Twitter. I’ve thought about taking a break for a couple years and decided over the holiday weekend that it was time to do it.
2
241
Part 1 of 3: I am going to take a break from Twitter. I’ve never really been much into it and I am going to focus my time elsewhere.
2
181
The evolution will be fascinating to watch! kdnuggets.com/2023/06/expect…
What Can We Expect From GPT-5? - KDnuggets

What Can We Expect From GPT-5? - KDnuggets

The moment we’ve all been waiting for - GPT-5 and the impressive capabilities of its predecessor, GPT-4.

kdnuggets.com
114
#Science for the day arstechnica.com/science/2023…
Liquid metal could turn everyday things like paper into smart objects

Liquid metal could turn everyday things like paper into smart objects

This futuristic new liquid-metal coating can make ordinary objects extraordinary.

arstechnica.com
104
It may sound a bit overly negative at first, but I think it is actually right on target. techxplore.com/news/2023-06-…
AI models feeding on AI data may face death spiral

AI models feeding on AI data may face death spiral

Large language models are generating verbal pollution that threatens to undermine the very data such models are trained on.

techxplore.com
68
I'm sure we can trust them and they'll never misuse all this personal data, right??? Technology is putting our freedoms at higher risk every day. datanami.com/2023/06/22/wher…
1
59
Some great examples with visuals of model collapse and how it works here bdtechtalks.com/2023/06/19/c…

ChatGPT will make the web toxic for its successors - TechTalks

Machine learning models trained on content created by generative AI will suffer from model collapse, according to a new study.

bdtechtalks.com
52
We're going to have some messes develop before we figure it all out venturebeat.com/ai/how-gener…
AI misuse reshapes security landscape | VentureBeat

AI misuse reshapes security landscape | VentureBeat

Why we must act now so that security professionals can develop strategies and react as large-scale issues arise with the evolution of AI.

venturebeat.com
64
The part I'm not sure of is how much you need people who like a given type of music to get a good read. techxplore.com/news/2023-06-…

Machine learning helps researchers identify hit songs with 97% accuracy

Every day, tens of thousands of songs are released. This constant stream of options makes it difficult for streaming services and radio stations to choose which songs to add to playlists. To find the...

techxplore.com
1
71
Pretty specific and yet still vague. Expect drama! arstechnica.com/information-…

Purely AI-generated songs declared ineligible for Grammy Awards

A work that contains no human authorship is not eligible in any categories."

arstechnica.com
2
2
118
Technology is allowing governments to do things they would be banned from doing directly. datanami.com/2023/06/22/wher…
1
62
More updates on #AI regulations coming venturebeat.com/ai/the-eu-ai…
EU and US race to regulate generative AI | VentureBeat

EU and US race to regulate generative AI | VentureBeat

Six months after ChatGPT became an overnight success, the US and the EU are racing to develop AI regulation.

venturebeat.com
1
58
Mimicking a person's voice or using it for training without permission is problematic. Creating a totally fake voice? Hard to argue against it. techxplore.com/news/2023-06-…
'Don't steal our voices': dubbing artists confront AI threat

'Don't steal our voices': dubbing artists confront AI threat

Voice actors around the globe are mobilizing against the unregulated use of artificial intelligence (AI) to generate and clone human voices that they fear poses a threat to their livelihoods.

techxplore.com
1
57
If this worked, it would be amazing discovermagazine.com/the-sci…
Will Artificial Intelligence Help Us Talk to Animals?

Will Artificial Intelligence Help Us Talk to Animals?

Artificial intelligence has made remarkable progress in recent years, but can it help us talk to animals? Explore the possibilities and limitations of AI in communicating with other species.

discovermagazine.com
51
Interesting that #ChatGPT will lie, deny it, then admit it when challenged further. Be careful using it and fact check carefully! blog.minethatdata.com/2023/0…
61
Sounds great, but will it work? forbes.com/sites/gilpress/20…
53
Good read. Training #AI on human created data is one thing, when models start training themselves on other model output, we'll drift from reality venturebeat.com/ai/the-ai-fe…
AI models risk collapse from self-training | VentureBeat

AI models risk collapse from self-training | VentureBeat

As a generative AI training model is exposed to more AI-generated data, it performs worse, producing more errors, leading to model collapse.

venturebeat.com
1
85
datafloq.com/read/taking-lar…
33
Interesting experiment and outcome. arstechnica.com/information-…
Researchers discover that ChatGPT prefers repeating 25 jokes over and over

Researchers discover that ChatGPT prefers repeating 25 jokes over and over

When tested, "Over 90% of 1,008 generated jokes were the same 25 jokes."

arstechnica.com
1
1
46
Load more