The Skeptic

Web browsing and cited sources baked into your Notion workspace

“The direct competitors here are Perplexity, which does cited web search better as a standalone, and ChatGPT with browse enabled, which already lives in more workflows than Notion ever will. The specific scenario where this collapses: any research task that requires more than five sources, real-time data accuracy, or a domain where citation freshness actually matters — Notion's model selection and crawl depth are opaque, and there's zero information on how often sources are verified. My 12-month kill prediction: OpenAI ships a tighter Notion-equivalent workspace integration and the marginal value of Research Mode evaporates, because the moat was convenience, not capability. To earn a ship, Notion needs to publish citation accuracy benchmarks and give users explicit control over source recency and domain filtering.”

Run Python & R code inside your search sessions, sandboxed and persistent

“Direct competitor is ChatGPT's Advanced Data Analysis — same concept, same tier pricing, and OpenAI shipped it first with broader file upload support. Perplexity's actual differentiator is that the interpreter is woven into a live web search session, so when you ask it to analyze current stock data or a just-published paper, the retrieval and the computation happen in one context window instead of you manually bridging two tools. Where it breaks: any workflow requiring external data sources beyond what the model can retrieve, complex multi-file projects, or users who need to reproduce work outside the Perplexity environment — there's no export-to-notebook story. What kills this in 12 months isn't OpenAI, it's Perplexity itself either commoditizing this into the free tier (making the $20 moat disappear) or getting acquired before the product matures. It wins if search-plus-compute becomes the default research workflow and Perplexity holds the search layer.”

RAG model with citation-level grounding for regulated enterprise search

“The direct competitors are Azure OpenAI with its own enterprise connectors, AWS Bedrock with Knowledge Bases, and Glean for the search-native buyers — Cohere is not in uncontested territory. Where this actually differentiates is that citation grounding is a model-level behavior, not a retrieval-layer trick: when the model declines to answer because the source doesn't support the claim, that's a compliance feature, not a UX quirk. The scenario where this breaks is any organization whose data lives outside the three supported connectors — if your source of truth is a custom ERP or a legacy SharePoint on-prem deployment, you're back to building pipelines. What kills this in 12 months isn't a competitor — it's that OpenAI and Anthropic are both racing to ship enterprise grounding natively, and Cohere's defensibility is deployment flexibility (on-prem, private cloud) that most of its target buyers haven't yet demanded.”

Extended thinking for grad-level math, science, and coding

“Direct competitor here is Gemini 2.5 Pro with thinking enabled and Anthropic's Claude 3.7 Sonnet extended thinking — o3 Pro is a legitimate participant in that race, not a pretender. The benchmark claims come from OpenAI's own evaluations, which should always be read as a floor not a ceiling, but the independent third-party evals on GPQA and competition math largely corroborate meaningful improvement over base o3. Where this breaks: anything requiring real-time data, multi-step tool use in complex agentic pipelines, or cost-sensitive workloads where the token budget for extended thinking makes it economically absurd at scale. The thing that kills this in 12 months isn't competition — it's OpenAI shipping o4 or o5 and making o3 Pro the mid-tier, which is exactly what they'll do. Ship it now if you have hard reasoning problems today.”