1/Large language models like Galactica and ChatGPT can spout nonsense in a confident, authoritative tone. This overconfidence - which reflects the data they’re trained on - makes them more likely to mislead.
371
2,3K
New updates just rolled out - joining a suite of Android features made to help you everyday! Open the thread to play a quick game of “This or That?”
17
327
Apple is reportedly preparing to allow third-party app stores on the iPhone https://t.co/ussIqPQkhd https://t.co/KVY5ddiqzG
36
238
Apple is reportedly preparing to allow third-party app stores on iOS https://t.co/UZXqGdqY9M https://t.co/9yQxpXq2qn
18
85
2/In contrast, real experts know when to sound confident, and when to let others know they’re at the boundaries of their knowledge. Experts know, and can describe, the boundaries of what they know.
20
386
Apple’s new whiteboard app gives you and up to 99 friends an "infinite canvas" https://t.co/zgGcQt36zd https://t.co/yoK389ePIc
16
127
Ever look at an API and want to know what artifact it belongs to or what the implementation looks like? Us too → https://t.co/JrBFjT5W5w https://t.co/qLhUg7RAQS
11
32
Prediction of Alzheimer’s improves with #MachineLearning = study shows which scans work best: https://t.co/nLQbrEftyf + The research paper: https://t.co/oRTXlfNPGh ———— #AI #PredictiveAnalytics #HealthTech #BigData #DataScience #IoT #IIoT #Biomarkers #DeepLearning #NeuralNetworks https://t.co/SxMXJ3shCo
6
33
I prepared my participation at #CES2023 with #ChatGPT🚀 Stunning.. See the result👉https://t.co/R9IvpIjm1U @sebbourguignon @enilev @Khulood_Almani @tobiaskintzel @PawlowskiMario @CurieuxExplorer @JeroenBartelse @FrRonconi @NevilleGaunt @chidambara09 @EvaSmartAI @mikeflache #AI https://t.co/2wrCANNuhU
15
21
3/Building large language models that can accurately decide when to be confident and when not to will reduce their risk of misinformation and build trust.
24
372
Logitech’s new iPad keyboard and stylus are a better buy than Apple’s https://t.co/acTCylxnrS https://t.co/Rji40olnQb
9
53
Instagram launches its BeReal clone, Candid Stories https://t.co/ftJ1G0TrB2 https://t.co/lVi1NkDqRU
10
21
New research released today indicates that Cuba has been using pieces of malware in its attacks that were certified, or given a seal of approval, by Microsoft. https://t.co/z9kkfFuIhM
9
17