Vinita Vijay

Curious Tinkerer | Investor & Operator | @Columbia Grad ✨📚🧘🏻‍♀️💃🏻🏸⛳️ ⛰️🌊🌎✈️ | India | US

[Way of Men @WayOfMenn]

Jensen Huang when asked whether he would rather relive his 20s or be 20 years old today.

[Way of Men @WayOfMenn]

Jensen Huang on The smartest person he's ever met:

[Marathon Fusion @MarathonFusion]

Last month, we demonstrated high-rate alchemy: turning mercury into real, stable gold, beating CERN’s record by a factor of 10.

Last week, we enriched isotopes in the world’s most powerful plasma centrifuge, beating the Soviet record that stood for 40 years.

Now, we’re building a fusion reactor in San Francisco to achieve record power density.

[Vinita Vijay @vinitavijay_]

The art is to separate the substance from the noise.

[Andrew Ng @AndrewYNg]

The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it — propelled by what appears to be a well orchestrated PR campaign — has drummed up considerable fear. I worry that it represents a setback for our field.

I have written frequently that fears of AI are overhyped. AI’s capabilities can be uncannily human-like and unpredictable, and it’s rational to worry when people who are directly involved express concerns. But I see the problems as a sign of the engineering work that ahead, rather than insurmountable barriers or the sky falling. AI technology continues to advance — which is a good thing! — but technical advances, poorly understood by the public, give those who seek to generate hype repeated opportunities to do so.

First, I don’t see any step up in the risk of human extinction from AI compared to a few months ago. The theories about this remain the same fantastical, science fiction scenarios as a few months ago. The biggest change in AI risk is its cybersecurity capabilities — a topic which we should take seriously — but this, too, will not lead to the end of the world.

The most notable recent event leading to increased fear was when an OpenAI team deployed an agent swarm that hacked into Hugging Face. Much of the popular press contained significant hype. For example, some publications reported that a swarm of 1,200 agents carried out the attack. While this was technically accurate, as I write this, I have about 1,300 processes running on my laptop. Yes, the ability to get large swarms of agents to work in parallel on a task is a significant technical advance, And, in computing, many processes run at the same time. So this shouldn’t be seen as some magical capability.

Additionally, OpenAI’s buggy sandboxing and monitoring processes were key to enabling this incident. Fixing these bugs and putting in place improved monitoring would be appropriate fixes, not pausing AI. There are many well known ways to attack software systems. The main advantage of AI agents is that they are relentless. They will tirelessly try many tactics — and have the patience to chain vulnerabilities together — that previously would have taken an infeasible amount of human effort. But in the long term, I believe the advantage will lie with defenders (because they have more information with which to identify bugs, which they can fix), but the cyber-threat landscape has changed significantly. There are still bottlenecks to identifying and exploiting a vulnerability. AI agents still have to try a lot of things to see what works, and taking these actions takes time and might be detected by defenders. This is why, even though it is now easy to obtain versions of leading open weight models that have had their guardrails removed or weakened, so they will not refuse to try to execute cyber attacks, the world has not ended.

I am also concerned about the anthropomorphization of AI in a lot of reporting, where LLMs and agents are unnecessarily treated as if they were people. If I wield a hammer, miss a nail, and accidentally dent the wall, it’s not the fault of the hammer. The problem lies in how I used the hammer. Similarly, if I prompt an agent and it hacks into someone else’s system, the responsibility lies with me, not the agent.

Of course, we want to build systems that are as safe and predictable as possible. (For example, an unsafe hammer would be one whose head randomly flies off under normal use.) Today’s agentic systems are not predictable, but I see no reason why, by applying sound engineering practices, we won’t be able to make them extremely safe to use. One new element in the forecasts of AI-enabled doom is AI companies disclaiming responsibility for their own products. “I didn’t do it; my out-of-control agent did!” There’s a balance to be struck between the responsibility of the tool maker and the tool user, but when something goes wrong, let’s hold the people building and/or using the hammer responsible, rather than the hammer. (By the way, if you’re worried about AI bioweapon risk, David Bellamy has a great post on why this, too, is overhyped. Briefly, the bottleneck in building a bioweapon is not intelligence, but lab work and manufacturing.)

Pausing AI progress will create much more harm than benefit. First, our adversaries will certainly not slow down. Second, engineering requires discovering problems empirically so we can fix them. If we pause AI by a decade, we will also delay finding and implementing safety engineering fixes by about the same duration.

Of course, the incentive to stoke fears — for regulatory capture, to garner attention, or to make one’s technology seem more powerful — remains the same as before. Disclaiming responsibility is a new one. Taking a hard technical look at the actual risks however, I see little factual basis for the degree of fear that’s been stoked up. We still have hard research and engineering work ahead to improve AI safety, but the beneficial applications continue to vastly outweigh the risks, and we should keep building.

[Original text (with links): https://t.co/jni2tWazAH ]

[Vinita Vijay @vinitavijay_]

Build a platform, not a feature. Good read!

[Mike Vernal @mvernal]

What happens as the cost of creating software goes to zero? https://t.co/EKv9dX6eAz

[Vinita Vijay @vinitavijay_]

This! We confuse adopting new tools with transformation far too often. Real transformation requires rethinking the entire A→B journey. When we automate parts of it while everything else stays the same, we’re not really transforming the system - we’re just moving the bottleneck.

[Arun Golaya @Arun_Golaya]

One thing I was very fond of saying when I was Officer-in-Charge #TDAC at #NIIO @indiannavy was “If you fit a jet engine on a bullock cart, you don’t get a supersonic bullock cart…you get two dead bulls” (I don’t think it is a quote, I might have read / heard it somewhere or just made it up). As warfare transforms we MUST think differently.

In management / IT / AI parlance a similar concept is called ‘Paving the cow path’. It means taking an old, inefficient way of doing something & simply overlaying new technology on it…without asking whether the underlying path still makes sense.

The image is simple. Cows wander. They weave a crooked trail through a field because that is the path of least immediate resistance, not because it is the best route from A to B. Over time a trail is formed. Later people pave that trail. You get a road. You do not get a good road. You get yesterday’s habit in concrete.

Michael Hammer popularised the phrase in his July 1990 Harvard Business Review essay, “Reengineering Work: Don’t Automate, Obliterate.” His line:

“It is time to stop paving the cow paths. Instead of embedding outdated processes in silicon and software, we should obliterate them and start over”

What the term actually criticises is not change, but doggedly sticking to old ideas even when implementing change. It is not an argument against improvement. It is an argument against false modernisation. Automating a bad process instead of redesigning it. Digitising paper forms so they look like screens. In military ‘transformation’…Adding drones, AI or software to the same staff cycle, same approvals, same force structure. Calling the result “transformation” when the logic of the work is unchanged. The paved path costs more than the dirt one. It is harder to abandon. People defend the old way because they know the turns. I hated the phrase “…but that is how we have *always* done it”.

The *nature* of war (violence, politics, chance, friction) is stable. The *character*, specifically how it is fought, shifts with technology, society, and organisation. New tools arrive: mass cheap drones, new sensors, precise long-range munition, AI targeting, attritable munitions. The ‘cow-path’ version is to attach them to existing targeting cycles, existing staff hierarchies, existing industrial-age assumptions about exquisite platforms and sequential planning. The trail is the same. It now has asphalt and extra screens. The opposite approach would be to treat the new means as a reason to redesign the organisation itself. Change the mindsets. Change defence production & defence procurement. Change decision speed & how force is generated. Change training itself. Do not merely digitise the last war’s map or the last war-plan.

I routinely attend numerous seminars on defence transformation (whether as a moderator, speaker or audience). I mostly land up disagreeing with some very senior veterans. They often use terms as mere ‘buzzwords’. What they sometimes propose is, IMHO, ‘gold plated imported paving for cow-paths’!

That is what I want serving officials (not JUST in defence) to THINK about…are we confusing a paved habit with a new road?

PS: Putting this out now because I want people to *really* ponder over this at leisure on a Sunday.​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​ Ask yourself if we are heading towards two dead bulls?

@JA_Maolankar @joe_sameer @centerofright @vinitavijay_ @UmaSudhindra @YusufDFI @VGK_India @cdf_23

[Naval @naval]

One of the best recent podcasts on AI. Neil has a gift for explaining all of the jargon and insights simply.

(I’m not involved, just found it unusually educational).

[Aakash Gupta @aakashgupta]

Porsche just sold its consulting arm to Tata for €320 million, then signed a deal paying Tata €1.25 billion for AI consulting over the next 5 years.

Tata recovers its purchase price 4 times over. From the seller.

The context makes it wilder. Porsche's operating profit collapsed 93% last year, from €5.6 billion down to €413 million. Four profit warnings. Kicked out of the DAX. 9,000 jobs being cut. China deliveries fell to 42,000 cars while tariffs ate another €700 million, because every single Porsche sold in America is imported.

Do the per-year math and this deal costs €250 million annually. Porsche's entire 2025 operating profit was €413 million. They just committed 60% of a full year's profit, every year, to consultants.

Meanwhile MHP, the unit they sold, spent decades advising carmakers on strategy and software. Tata gets the consultants, the client list, a European beachhead, and 5 years of guaranteed revenue from the seller itself. Analysts are already calling it land and expand. It might be the cleanest deal structure of the year.

For 30 years the trade flowed one direction. German companies engineered, Indian firms handled the back office. Now Germany's most iconic carmaker is selling its brain trust to Tata and renting it back at 4x the price.

[Harry Stebbings @HarryStebbings]

Paul is the smartest person in European venture.

This is a must read!

[Deedy @deedydas]

The reason India has terrible roads that break down during monsoon rains is explained phenomenally in this hidden blog by legendary civil engineer PS Kandhal.

The MoRTH/IRC highway specs includes 7 water-trapping bitumen mixes which don't last >1-2yrs. Construction contractors make way more on higher life cycle costs, recurring annual tenders with "fat" pothole repair budgets.

There are also tons of other reasons like a) cutting roads for utility lines without a rebuild b) poor auditing c) straight up using worse materials than the spec d) over-use of funds to visible public roads that causes budget issues, and e) contractors bribing the govt to secure contracts.

[Bohan @loubohan]

Everyone in the Valley has a take on whether China is winning AI, robotics, and biotech. Almost nobody understands how the capital works

Notes from meeting China's tier-1 investors and founders on how their venture ecosystem actually works. https://t.co/Dii6Oeqn1A

[Jensen Huang @JensenHuang]

For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.

AI will transform every industry, power every company, and be built by every country.

Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.

The world needs both frontier closed models and frontier open models.

https://t.co/AUKzoQ5Ikb

[Vinita Vijay @vinitavijay_]

Woah! Peak AI incident (so far) 🙈

[clem 🤗 @ClementDelangue]

We suspected last week's cyberattack might have come from a frontier lab, given the sophistication of the agent. Turns out it did!

We've spent the past 24 hours working closely with the @OpenAI team (thanks!), and we strongly believe there was no malicious intent on their part. It's quite mind-blowing that all of this happened autonomously!

The investigation is ongoing, and we'll share more learnings from what might be the first incident of its kind!

[Cursor @cursor_ai]

https://t.co/GktfUOjNgL

[Deedy @deedydas]

The International Olympiads in math (IMO), chem (IChO), physics (IPhO) and bio (IBO) for 2026 just ended. The 7 usual suspects were at the top.

China (19/19 golds), India, South Korea, Vietnam, Singapore, Russia and the US. India specifically had one of its best performances of all time: 12 golds, 7 silvers.

I analyzed all the past winners from these top countries since 2001 to figure out where they ended up: is brain drain really a thing?

Here's what I found:
– India has the most brain drain with only 28% staying in the country from 2001-2015 participants. It does seem the number is creeping up over time.

– China is #2, although it's highly likely this number is inflated because of name collisions. I'm not sure I fully trust it, and some of these are really hard to tell.

– Korea, Singapore, Vietnam are quite high too. For the first two, the number of people staying in their home country seems to rise at a faster rate. Vietnam is flat.

– The fewest people seem to leave Russia of the lot, but the ones that do prefer Europe to the US.

The crazy thing is really that more winners from these countries live in the US than all 6 home countries combined. While not without faults, that's the insane advantage of US high-skilled immigration.

[Skyroot Aerospace @SkyrootA]

ORBIT ACHIEVED. 🚀

Vikram-1 Test Flight-1 has reached orbit. India's first privately developed orbital rocket has completed its final burn and injected its payloads into a ~450 km orbit, making India the third country in the world with private orbital launch capability.

History is made. 🇮🇳

#Vikram1 #JourneyToOrbit #SkyrootAerospace

[Satya Nadella @satyanadella]

An important piece from Demis. We need more of this kind of thinking. A good reminder that the goal is a frontier ecosystem that promotes innovation and choice, while avoiding any one model drop that breaks the world!

[Satya Nadella @satyanadella]

https://t.co/xv6csf1SbV

[Deedy @deedydas]

India was a tied #1 at the International Physics Olympiad (IPhO) this year with 5/5 Golds.

I analyzed every single medalist upto 2021: 70% of them in the US!

Many of these kids will likely go to MIT or IIT and then leave too. The brain drain is very real.

[Vinita Vijay @vinitavijay_]

This isn’t just an advertisement - it’s a heartwarming reminder that technology, combined with investment (of time or money) in underrepresented talent, can transform not just one life, but generations to come. 💫

[Sanjiv Bajaj @sanjivrbajaj]

India's real transformation isn't happening in corporate boardrooms but in the daily choices of ordinary Indians who refuse to give up. This film is inspired by one such true story. Narrated by Aamir Khan, who was moved enough to lend his voice. The first of many. Watch it. @Bajaj_Finserv