Not sure why indices are down today. My momentum portfolio is up 0.8% (thanks Tutor Perini) and my divi Portfolio is up 0.35% (thanks Cummins and Merck).
1 Like
nVidia will either fix it tomorrow OR itâll start tumbling down. Or not. Who cares, divvies soon 
2 Likes
But, Sir, the divvies are why the market dropped!
Since its recent top, about 2% of divvies were payed out, hence the market dropped accordingly.
Donât you understand?
</sarcasm>
1 Like
I bought a lot today. The main reason is because I sold a big position last week though
No liquidity here, payday in a week.
Canât wait for the market bump, thanks for the insider advice.
3 Likes
Last week I attended the biggest pharma/health economics conference of the year. One takeaway regarding use of AI in this field is that in terms of use cases Iâm still not seeing anything groundbreaking, and weâre three years in. Almost everyone has some solution but itâs mostly LLMs, text generation, data management, assisting with economic modelling. Besides modelling AI just helps do what more junior people usually do, like in law.
Both manufacturers and their customers (healthcare system payers) are slowly moving towards acceptance of AI-generated content but itâs a few years away from âa few clicks do 12 months of workâ. Itâs moving but itâs slow.
The human is still firmly in the loop, and I believe itâs unlikely that weâll be kicked out soon.
Moreover it (AI) doesnât and wonât replace human relations in terms of making business happen.
6 Likes
When I read the news, I often see concrete examples in pharma â things like AI-driven drug discovery, reduced development times (e.g., DSP-0038), process optimization, tablet inspection and quality assurance (around 95% accuracy at 1,000 capsules per minute), the use of AI in identifying adverse drug reactions, or tools developed by companies like Owkin. But since Iâm not from the pharmaceutical industry, I probably have fewer insights than you.
At the same time, finance is clearly shifting as well: companies like OpenAI are hiring highly paid bankers from Goldman Sachs and others to help craft prompts for time-intensive and therefore cost-intensive processes such as IPOs. Setting aside the irony that these people may effectively be digging their own professional graves, the compensation seems to be extremely high. The IPOs are just one field to mention.
In my view, both pharma and finance have moved well beyond the âearly experimentationâ phase â weâre now entering a period of concrete implementation. And I honestly think that within the next five years we will see a significantly negative shift in the job market as these technologies become more deeply embedded.
1 Like
This for sure, big advances in holy grails like protein folding, molecular docking. I am paying attention but just out of personal curiosity as drug design was part of my PhD. The scientific and applicable advances are indeed happening, itâs just that I am not in this field.
Yeah, I agree 100%. What I am seeing is that AI can and does/will help in dealing with very large bits of repetitive, structured text/tables/data. I bet they may be hiring lawyers too (remember reading one of the first tests for LLMs was pitting them against the work junior associates would do, LLMs blew the junior lawyers out the water, and that was 2 years ago), and for sure some people may be digging someoneâs professional grave, could be their own.
1 Like
Thanks for the thought, Iâm too far removed from this though, last time I did anything really scientific was in 2011. Pre-clinical is not even the biggest hurdle/cost of development, but it certainly is very interesting. Whatâs needed is lateral thinking, making connections where they donât seem plausible/probable. That or just throwing computing power at the problem! About 20 years ago I visited GSKâs Stevenage plant as part of my MSc, they had multiple floors filled with robots doingâŠhigh-throughput screening for active compounds. Thatâs prime AI space, not doing the screening at all, or helping go from 100,000 compounds to be tested in a lab down to 100.
Exactly, instead of doing with robots, trying to simulate this in software instead. A big time and money saver.
1 Like
Not trying to time anything 
⊠but could it be the ânext big AI-related themeâ? Iâve checked some Biotech ETFs (eg. BTEC) and last few months it was going straight up (OK, like most of the things 
)
Biotech were down because of high interest rates. Things have been picking up because of gradual IR reduction but more importantly acquisitions. Acquisitions are 9 out of 10âthe hopeful end of those biotech. Levels are still lower than few years back but some optimistic has returned.
however, some recent IPOs of biotech were above sensible fair value and thus struggle.
I think biotech will bounce back but wonât be straight line. Lowering rates will help but broader market sentiment will pull things down across so different factors at play
1 Like
Show must go on tonight.
1 Like
I wonder if you have a source on that. It doesnât quite match what I believe is going on, but is close enough that it could be a garbled version of what was described in the October 21st issue of Money Stuff.
Basically what theyâre doing is not hiring ex-bankers to do prompting, but hiring them in to create very high quality training materials. So basically do exactly the same job youâd do as a junior investment banker with writing reports and making presentations based on source material. And then obviously models are trained or tuned with this data.
Crucially, you donât need to hire senior bankers for this, because thatâs not the job theyâre looking to automate right now. The people who did a junior career in investment banking but left for whatever reason will do. The reported rates for these gigs are $150/hour. Good money for a low stress job (theyâre fake reports, how stressful can it be?!), but not for people with active Wall Street careers.
2 Likes
Let me guess, they beat, raise guidance and then the market tanks anyway?
1 Like
Sentiment is very bearish in my window to the world (reddit), that theyâll barely hit or even miss projections, and market will tank hard. This makes me think itâll go the other way in fact. That said, their earnings vs forecasts have been edging towards a plateau the last year - if I remember correctly - so barely hitting could be what happens.
To be honest I donât care much!
The problem is that it is impossible to guess what will happen. And even worse, even if we know what will happen, it is impossible to know how the market will react to it!
1 Like
We know what the expectations are:
Options markets imply traders are expecting the AI chipmakerâs share price to move by an average of 6.4 per cent in either direction when markets open on Thursday, equivalent to a gain or loss of about $280bn in value
Source: Client Challenge
1 Like
This is evident from price action in Bitcoin. Almost everything crypto world wanted from US govt is happening, but price is moving in another direction.
itâs tough to know market action when prices are driven by so many unknown variables
3 Likes