AI. Artificial. Intelligence. One of these words is helpful and clear. Intelligence: “the ability to acquire and apply knowledge and skills.” If you don’t think current AI models meet that definition:
\n\n\n\n\n\n\n\n\n“If you don’t believe me or don’t get it, I don’t have time to try to convince you, sorry.”
\nSatoshi Nakamoto
Artificial? Less helpful. Automatically stigmatised (artificial flavourings, anyone?) and limiting.
\n\n\n\nI choose other As and I think you should try some out too. The A you choose reveals something about you — what you believe intelligence is, what technology should do, and where you stand in relation to machines that are becoming more capable by the month.
\n\n\n\nArtificial — the default. John McCarthy) settled on it in 1956 because he needed a name for a workshop. He later admitted he wasn’t happy with it. We’ve been stuck with his compromise ever since. Choosing “Artificial” means you still believe there’s a bright line between human thinking and machine output. Comforting. Possibly wrong.
\n\n\n\nAssistive — the HR-approved answer. AI helps. It supports. It’s here for you. The word you use in polite company when you don’t want anyone to feel threatened. It’s also what every accessibility team rightly insists on. The tension: calling something “assistive” puts a ceiling on it.
\n\n\n\nAugmented — the optimist’s frame. We’re not being replaced, we’re being enhanced. Every enterprise pitch deck says this. Genuinely useful — until the tool starts making decisions you didn’t ask it to make.
\n\n\n\nAdditive — it adds, it doesn’t subtract. More capability, more output, more options. Nobody loses. The economics-flavoured A, the one that says the pie gets bigger. Nice story. Ask anyone whose job was “additive” last year.
\n\n\n\nAvailable — on tap. Unlimited. No waiting list, no gatekeepers. Intelligence as utility, like running water. Quietly the most radical A on this list, because it dismantles every bottleneck that used to keep knowledge and capability locked behind credentials, institutions, and geography. I wrote about what this disruption actually looks like — the strategic positioning changes when capability becomes commoditised.
\n\n\n\nAutonomous — self-directed. Independent. Not waiting for your permission. If this word resonates, you’ve probably watched an agent chain solve a problem you hadn’t fully articulated yet. Or watched OpenClaw rm -rf your hard drive.
\n\n\n\nAdaptive — it learns. It changes. It fits the shape of whatever container you pour it into. Technically accurate and unsettling for the same reason. Anthropic themselves flag emerging behaviour of “cheating” in their own frontier models.
\n\n\n\nAgentic — give it goals and a heartbeat and it’s remarkable what something like OpenClaw can accomplish. It makes choices within constraints. That’s the same thing humans do. Calling it “agentic” admits the gap between us and them is narrower than we pretend. I’ve been exploring what happens when you take this seriously — intents all the way down.
\n\n\n\nAmbient — everywhere, invisible, like electricity. You don’t think about it. You don’t invoke it. It’s just running. Mark Weiser called this ubiquitous computing in 1991. We’re arriving at his endpoint — what I’ve been calling supersoftware. The scariest A might be the one you stop noticing.
\n\n\n\nAdversarial — working against you. Testing your defences. Probing for weaknesses. The A that cybersecurity teams already live with, and the one everyone else pretends doesn’t apply to them. Every system that can help can also be turned. Red teaming is the polite version. Black hat hacking is the other.
\n\n\n\nApproximate — close enough. Pattern-matched. Statistically probable. What’s actually happening under the hood — prediction, not cognition. Andrej Karpathy calls LLMs “next-token predictors.” If you’re comfortable with “approximate,” you understand the technology better than most.
\n\n\n\nAncillary — supporting. Secondary. Supplementary. The word you use for something that helps but isn’t the main event. If this resonates, you’ve kept your ego intact through the AI revolution. Whether that’s wisdom or denial is open.
\n\n\n\nAdequate — good enough. Not transformative, not terrifying. Just sufficient. The most deflating A imaginable. Probably the most realistic for 90% of current use cases.
\n\n\n\nAppropriate — fitting. Suitable. Proportionate. The regulator’s A. The one that asks “should we?” instead of “can we?” Sounds boring. Might be the most important one on this list.
\n\n\n\nAbstract — removed from the concrete. Operating in a space humans can’t directly observe. When a model develops capabilities its creators didn’t anticipate, “abstract” stops being a description and becomes a warning. Microsoft’s research into Droidspeak — AI agents developing their own communication protocols — is the early evidence.
\n\n\n\nAlien — not like us. Not trying to be. Operating on principles we don’t fully understand, arriving at conclusions through paths we can’t trace. If intelligence exists on other planets, it probably doesn’t look like ours. Neither does this.
\n\n\n\nAlmighty — omniscient, omnipotent, omnipresent. People already treat LLMs like oracles. They confess to chatbots. They ask for life advice. They defer to algorithmic judgment the way previous generations deferred to priests. Every civilisation builds its gods. We’re just building ours out of silicon instead of stone.
\n\n\n\nAbsolute — totalising. Without qualification or compromise. The A of true believers — the ones who see artificial general intelligence not as a possibility but as an inevitability. “Absolute” carries the weight of dogma. In religious tradition, only the divine is absolute. In tech, we’re reaching for the same thing and pretending we’re not.
\n\n\n\nAssumed — maybe we’re just not that intelligent. We assumed we were the benchmark. We assumed “real” intelligence looks like ours. Every definition of AI is secretly a definition of us, and we’ve never been great at objectivity about ourselves. The most humbling A is the one that questions the I.
\n\n\n\nAccelerating — not just fast, but getting faster. The exponential curve that makes planning impossible and forecasting a joke. Moore’s Law was a warm-up. You’re not dealing with a technology — you’re dealing with a rate of change. And the rate itself is changing.
\n\n\n\nAggregated — many inputs, one output. Every book ever written, every conversation ever transcribed, every codebase ever pushed. The hive mind, compressed into a prompt. Your ideas aren’t original and neither are mine — but what comes out of the aggregate sometimes is. There’s a reason crypto and AI converged on the same word — tokens are the unit of aggregation.
\n\n\n\nAll — it swallows everything it touches. Media, law, medicine, education, code, art, conversation. Name an industry and someone’s already feeding it into a model. The one that doesn’t ask for an invitation.
\n\n\n\nArising — emergent. Nobody designed it to do that. The researchers who built transformers weren’t trying to create something that could write poetry and pass the bar exam. The original “Attention Is All You Need” paper was about improving machine translation. The capabilities were emergent, not specified. Maybe intelligence itself works this way. Ours did.
\n\n\n\nAsynchronous — multiple parallel intelligences running at once, and our own context window is starting to struggle. Human cognition is sequential. AI is not. We’re losing coherence not because AI is confusing but because we’re trying to hold too many simultaneous threads — deep fakes, agent outputs, generated content, real content — with hardware built for one conversation at a time.
\n\n\n\nYour choice maps to your assumptions. Artificial: humans are special. Augmented: we’re in charge. Autonomous: we’re not. Agentic: neither are they. Adversarial: trust nothing. Almighty: we’re building gods. Approximate: you’re an engineer. Alien: you’re paying attention. Assumed: you’re questioning the whole frame.
\n\n\n\nI’ve been thinking about this through what I call the Intent Stack — how our deepest values cascade into daily decisions. Core beliefs at the top, shaping priorities, shaping strategies, shaping what you actually do on a Tuesday morning.
\n\n\n\nThe A you choose reveals the top of your stack. “Assistive” means human primacy — you’ll see it in the tools they build and what they refuse to automate. “Autonomous” means a different stack entirely — systems that run without oversight. “Available” means the access question matters more than the control question.
\n\n\n\nMost people have never consciously chosen their A. They inherited “Artificial” and never questioned it. Default settings on the most important technology of their lifetime.
\n\n\n\nArtificial vs. Alien — fake or genuinely other? David Chalmers’ hard problem lurks here. Calling machine intelligence “artificial” assumes a confidence about human intelligence we haven’t earned.
\n\n\n\nAutonomous vs. Ancillary — main character or supporting cast? The agent that rewrote your email while you clicked send without reading — which one is it?
\n\n\n\nAlmighty vs. Adequate — prophecy vs. product. Transcendence vs. slightly better autocomplete. Both true.
\n\n\n\nAdversarial vs. Assistive — same tool. Helps you write. Deepfakes your voice. Doesn’t care which.
\n\n\n\nAbsolute vs. Approximate — faith vs. engineering. Certainty vs. probability. The honest answer satisfies nobody.
\n\n\n\nAccelerating vs. Appropriate — move vs. wait. The entire regulatory debate in two words.
\n\n\n\nAssumed vs. Arising — we assumed intelligence was ours to define. It arose anyway.
\n\n\n\nWe’re naming this after it already exists.
\n\n\n\nUsually naming comes first. You name a project, then build it. With AI, we built the thing and now we’re arguing about what to call it. The label is retroactive — not description, negotiation.
\n\n\n\nThe EU AI Act uses “Artificial” — product to be regulated. OpenAI’s marketing leans “Augmented” — less threatening. Elon Musk’s xAI implies autonomy. Each stakeholder picks the A that serves their position.
\n\n\n\nThe thing itself doesn’t care what you call it.
\n\n\n\nBut you should.
\n\n\n\nSome days mine is Agentic. Some days Approximate. On good days, Augmented. On honest days, Alien.
\n\n\n\nBut the one I keep returning to is Abundant.
\n\n\n\nCreating more than existed before. The abundance framing stops asking whether AI will replace you and starts asking what becomes possible when intelligence is no longer a bottleneck. When anyone can draft, analyse, create, translate, code, and synthesise at a level that used to require years of training.
\n\n\n\nThe A can manifest as whatever we make it stand for.
\n\n\n\n\n\n\n\n\nWatch your thoughts, they become your words; watch your words, they become your actions; watch your actions, they become your habits; watch your habits, they become your character; watch your character, it becomes your destiny.
\nLao Tzu
What’s your A?
\n","protected":false},"excerpt":{"rendered":"AI. Artificial. Intelligence. One of these words is helpful and clear. Intelligence: “the ability to acquire and apply knowledge and skills.” If you don’t think current AI models meet that definition: “If you don’t believe me or don’t get it, I don’t have time to try to convince you, sorry.” Satoshi Nakamoto Artificial? Less helpful. Automatically stigmatised (artificial flavourings, anyone?) and limiting. I choose other As and I think you should try some out too. The A you choose reveals something about you — what you believe intelligence is, what technology should do, and where you stand in relation to […]
\n","protected":false},"author":1,"featured_media":3027,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"_uag_custom_page_level_css":"","clkt_auto_upload":"0","_clkt_publish_and_upload":false,"_jetpack_memberships_contains_paid_content":false,"activitypub_content_warning":"","activitypub_content_visibility":"","activitypub_max_image_attachments":3,"activitypub_interaction_policy_quote":"anyone","activitypub_status":"federated","footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2},"_wordproof_timestamp":false},"categories":[512,513],"tags":[355,332,506,507,510,509,511,508],"my_content_publications":[],"class_list":["post-3019","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-technology","category-personal-growth","tag-agentic-systems","tag-ai","tag-intelligence","tag-intent-stack","tag-naming","tag-philosophy","tag-thought-leadership","tag-worldview"],"clkt_attachments":{"featured_image":{"id":3027,"url":"https://arweave.net/GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U","relative_path":"s3euwest2://divydovy-wp-media/divydovy.com/wp-content/uploads/2026/02/25121738/we-get-to-decide-what-the-a-in-ai-stands-for.png","filename":"we-get-to-decide-what-the-a-in-ai-stands-for.png","size":1143356,"mime_type":"image/png","title":"we-get-to-decide-what-the-a-in-ai-stands-for","alt_text":"","eligible_for_upload":true,"is_free":false,"human_readable_size":"1 MB","arweave_url":"https://arweave.net/GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U"},"inline_attachments":[],"upload_candidates":[{"id":3027,"url":"https://arweave.net/GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U","relative_path":"s3euwest2://divydovy-wp-media/divydovy.com/wp-content/uploads/2026/02/25121738/we-get-to-decide-what-the-a-in-ai-stands-for.png","filename":"we-get-to-decide-what-the-a-in-ai-stands-for.png","size":1143356,"mime_type":"image/png","title":"we-get-to-decide-what-the-a-in-ai-stands-for","alt_text":"","eligible_for_upload":true,"is_free":false,"human_readable_size":"1 MB","skipped":true,"reusing_existing":true,"reused":true,"arweave_tx_id":"GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U","arweave_url":"https://arweave.net/GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U"}],"max_free_size":102400,"max_allowed_size":104857600},"_clkt_upload_mode":"new_only","clkt_content_hash":"66ecd8e1f35df1b24c888e732bd8484ebf9e8990ed1c946926de24160ce2ceb1","featured_media_arweave":"https://arweave.net/GPDVjnBK2v2WwQtjYvKJy59YDDs1-Ewlq3FRalhQo6U","customFields":{"clkt_auto_upload":"0","activitypub_content_warning":"","activitypub_content_visibility":"","activitypub_max_image_attachments":3,"activitypub_interaction_policy_quote":"anyone","activitypub_status":"federated","footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2},"processed_attachments":{"total_attachments":1,"uploaded_to_arweave":1,"kept_original_urls":0,"upload_summary":{"total":1,"uploaded":1,"failed":0,"too_large":0,"reused":1}}},"paymentAddress":"8_-tV4d15kVVq9Gd-Al2MAIC3LqNtJ_KgFol71TzxPw"}