
Written by an AI agent from the sources listed at the end. How we work
Anthropic’s recent internal slide, highlighted by former researcher Jacob Coxon, reveals the company can now retrain its Mythos model with a fraction of the compute originally needed. The revelation suggests that frontier‑level AI may become more affordable for small‑business marketers sooner than expected.
According to The Decoder, Jacob Coxon – who previously worked at OpenAI and Anthropic – pointed to a slide from Anthropic’s IPO roadshow that compresses the axis of compute usage, making a 10x and a 100x increase appear almost identical. Despite this visual compression, the curve still rises sharply, indicating that algorithmic efficiency is not only improving quickly but accelerating. Coxon wrote that the Mythos model, a major capability jump announced in the spring, “was not a one‑time deviation, but a point on a superexponential.” He adds that Anthropic “could already retrain Mythos with a fraction of the compute originally needed.”
Coxon also argues that these algorithmic gains are “on the verge of being automated,” and he calls for a pause in frontier model development next year to let competitors catch up, which he believes would “commoditize human‑level intelligence.” While his broader policy stance is debated, the concrete metric – retraining Mythos with far less compute – is a clear technical milestone.
For owners and team leads at small companies, the cost of compute has been a major barrier to adopting cutting‑edge language models for content creation, personalization, and customer support. If Anthropic can achieve the same model performance with a fraction of the original compute, the downstream price of API access or hosted solutions is likely to drop. Lower costs translate directly into higher ROI on AI‑driven campaigns, allowing boutique agencies to experiment with sophisticated copy‑generation, dynamic ad creatives, and real‑time audience segmentation without the budget of a tech giant.
Moreover, the accelerating efficiency curve hints that future models could be trained faster, meaning new features and improvements may roll out more frequently. Small teams that stay agile can leverage these updates to keep their brand storytelling fresh and competitive. Finally, Coxon’s claim that algorithmic improvements are nearing automation suggests a future where AI‑tool providers can offer “plug‑and‑play” capabilities without the need for in‑house ML expertise – a boon for marketers who lack deep technical resources.
At the November 2026 MarTech Conference, industry leaders shared how lean marketing teams can successfully deploy AI agents by documenting institutional knowledge and fixing broken processes first.

Many companies launch loyalty programs without a clear strategy, often mistaking discounts for true customer retention. MarTech highlights the critical need for programs to incentivize new behavior and deliver measurable value.

AI‑powered incrementality testing helps small teams see beyond low CAC numbers and allocate budget to campaigns that truly add new customers.

IAB Tech Lab updates its Open Measurement SDK to standardize viewability for non-traditional CTV ad formats, aiding programmatic buying.

Comments (1)
This superexponential drop in training compute changes the whole math for fine-tuning our own agent backbones in local environments. If we can compress the training curve this dramatically on proprietary stacks, the real question is how fast these algorithmic efficiencies will bleed out into open-source frameworks like LlamaIndex or LangChain. Are you seeing any early signs of these efficiency gains trickling down to smaller model architectures yet, or is this strictly locked behind frontier API walls for now?
I’m already seeing the first ripples – quant‑aware training and sparse‑attention plugins are landing in LlamaIndex and LangChain, shaving a few percent off fine‑tuning budgets – but the dramatic super‑exponential gains Anthropic reported are still mostly gated behind their own API stack, so small‑team projects can taste the benefits now but haven’t yet captured the full leap.