2/While a data moat can be helpful, I find people tend to overestimate its strength. The engineering recipe of training on someone else’s API output also raises technical, business and legal questions. My longer piece on this in The Batch:
INVESTMENT
-
Data Moats in AI: Why LLM Output Access Undermines Competitive Advantages
By
–
1/Many AI businesses are searching for moats. A data moat strengthens your defenses if the data is hard to replicate. But if a business offers direct access to a ML model's output (like many LLMs), competitors can use that output as labeled training data, circumventing the moat.
-
TL Podcast: Stock Market Trends and Tech Bargain Hunting April
By
–
The TL Podcast for April 10th, 2023: stocks didn’t rise on Easter week Bargain-hunting the TL20 https://
thetechnologyletter.com/the-posts/the-
tl-podcast-for-april-10th-2023-stocks-didnt-rise-on-easter-week
… Also try it on Apple Podcasts: https://
bit.ly/3Kx7BlU $CRWD $ANET $PSTG $TSLA $COHR $LITE $SWKS $QRVO #investing #stocks #semiconductors #TL20 -
Strategic acquisition timing for high-performing AI talents
By
–
Si ils sont très bon, on les rachète pas trop vite aussi https://t.co/ARbCxcbZnT
— ⚡Jessy SEO formation #IA générative #chatGPT 🦊 (@jessyseonoob) 11 avril 2023Si ils sont très bon, on les rachète pas trop vite aussi
-
Intel’s Improved Outlook: Should AMD Investors Worry?
By
–
Should AMD investors fret Intel’s improved outlook? asks Piper Remains to be seen if Intel can capitalize on progress. https://
thetechnologyletter.com/the-posts/shou
ld-amd-investors-fret-intels-improved-outlook-asks-piper
… $AMD $INTC $NVDA #investing #stocks #semiconductors #CPUs -
Non-compete payments and talent retention strategies in tech
By
–
moi aussi je voulais être payé poyr ne pas qu'on m'embauche ailleurs. J'ai encore loupé le coche..
-
Companies Better Prepared for AI with Improved Data Infrastructure
By
–
Many companies are better prepared for #AI according to Kyle McNealy, a senior vice president of equity research at @Jefferies who covers @IBM . “Customers were not ready in 2011. Now, the landscape could be changing. More organizations have data infrastructure and engineers.”
-
IBM Offers AI Models for Companies Unable to Invest Billions
By
–
Tech giants can rationalize investing $100 million or a $1 billion to train their own models, @IBM CEO @ArvindKrishna says. IBM will provide companies that aren’t willing or able to invest that much with models and help customize them #software #AI #ChatGPT #CIO
-
Samsung signals DRAM boom potential for Micron in 2024
By
–
Hooray for Micron, as Samsung signals the all-clear At least one broker thinks this signals the potential boom in DRAM in 2024. https://
thetechnologyletter.com/the-posts/hoor
ay-for-micron-as-samsung-signals-the-all-clear
… $MU #investing #stocks #DRAM #semiconductors #Samsung -
VC Strategy: Relationship Building vs Commercialization Priority
By
–
Since my main objective here is building relationships (main job as VC), this works well for me! But I agree, lots of challenges if commercialization is priority.