It took 20 years for each internet user to consume 50Gb of memory annually. Could AI 5x that in just 5 years?

We're witnessing a massive shift in AI, from merely assisting with work to doing the work itself.

While the previous generation of technology focused on assistance through autocomplete, suggestions, and drafts, the new paradigm centers on agentic systems that execute tasks end-to-end. In coding, this transition is already measurable: Claude Code downloads soared 70x in under a year, while Codex installations rose 7x in seven months.

The data shows a sharp acceleration in early 2026, as new models unlocked the “agentic” era. Coding is the first vertical where this pattern is undeniably clear, and it serves as a blueprint for what we expect to see across industries.

For more on this C:\Take, watch Lucas:

The SaaS model is being rewritten in real-time.

As AI labs outscale some of the most iconic software businesses in history, we are witnessing a fundamental shift in the unit of economic value: the transition from selling software (per-seat) to selling work (per-output).

By shifting the unit of value from the tool to the results, the addressable market potential expands by 25x. We are moving from a $0.2T software market to a $5.5T "services-as-software" paradigm.

For more on this C:\Take, watch Lucas:

Coincidence or catalyst?

AI/ML roles are surging while SWE stays flat. It feels like a hiring reallocation, not a cycle.

In just 6 years, Supabase went from zero to 7M+ users.

The real story? 80%+ of that growth was just from the last year alone. From January 2025 to today, Supabase added ~6M+ new users driven by their acceleration to support developers in the AI supercycle.

A (new) star is born.

The new WFH is WFC (work from compute).

AI will undoubtedly eliminate some tasks, but history suggests it will also create entirely new opportunities.

We looked past the AI “doomer” headlines to examine the history of creative destruction. From 1910 to 1970, the agriculture segment shed more than 8m jobs, while employment in newer industries grew by 46m over the same period.

In 1973, the New York Times predicted ATMs would cut bank teller jobs by 75%. Instead, teller employment grew by 81% from 1970 to 1988. After spreadsheets emerged in the late 1970s, bookkeeping roles declined, while accounting surged and modern financial analysis rose to prominence.

In fact, 60% of the jobs that exist today did not exist in 1940 — not in spite of technological change, but because of it.

For more on this C:\Take, watch Max:

Incredible revenue ramp at scale.

The rate of innovation is faster than anything we’ve seen before!

The ATM Paradox.