{"id":13914,"date":"2026-08-18T05:12:05","date_gmt":"2026-08-18T05:12:05","guid":{"rendered":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/"},"modified":"2026-08-18T06:41:39","modified_gmt":"2026-08-18T06:41:39","slug":"how-to-use-ai-at-work-without-sounding-like-ai-2","status":"publish","type":"post","link":"https:\/\/www.exporis.ch\/it\/how-to-use-ai-at-work-without-sounding-like-ai-2\/","title":{"rendered":"How To Use AI At Work Without Sounding Like AI"},"content":{"rendered":"

Most marketing teams already use AI somewhere in the writing process. It can organise research, turn rough notes into an outline, compare headline options, summarise interviews or give you a first version of something you would otherwise have to start from a blank page. When deadlines are tight and the same team is producing website articles, newsletters, social posts and client material, that kind of help is hard to dismiss.<\/p>\n

You usually notice the problem later, when you read the finished article back and realise it could have been written for almost any company. The grammar is fine, nothing is obviously wrong, but the language keeps circling around familiar ideas about changing expectations, meaningful engagement, new opportunities or an increasingly complex market. Paragraphs arrive in roughly the same shape and length, while every few sentences the copy pauses to explain why the previous point is important.<\/p>\n

People who work with AI regularly are getting better at hearing these patterns. There are now tools trying to detect them as well. GPTZero and Pangram analyse characteristics of the writing itself, while Anthropic has taken a different route with newer Claude models, which can add an invisible machine-readable watermark to generated text. That watermark may later indicate that Claude was involved in producing a passage even when the language no longer looks obviously machine-generated.<\/p>\n

If you use AI professionally, you do not need to pretend that it played no part in the work. It is more useful to decide what you are happy to delegate and what you still want the person whose name appears on the article to contribute.<\/p>\n

Start with information the model does not already have<\/h2>\n

A broad prompt gives the model very little to work with.<\/p>\n

Ask for a thought-leadership article about customer loyalty in financial services and you can make a reasonable guess at what will appear: trust, personalisation, digital expectations, customer experience and perhaps a paragraph about balancing technology with the human touch.<\/p>\n

Those subjects belong in financial-services marketing, but they are not an argument by themselves. An insurer, private bank and fintech could all publish variations of the same article.<\/p>\n

Someone inside the company probably has more useful information. Sales colleagues may keep hearing one objection before prospects sign. Clienti<\/a> might ignore a feature that management considered central to the proposition. A campaign could have delivered impressive traffic but poor-quality leads. Analytics may show that applicants repeatedly leave the same page of an onboarding process.<\/p>\n

Put those things into the prompt.<\/p>\n

\u201cClients increasingly expect personalised services\u201d gives the model plenty of room to reproduce familiar marketing language. \u201cProspective clients repeatedly asked advisers about withdrawal restrictions before asking about expected returns\u201d gives you something much more useful to examine. Perhaps liquidity is influencing the decision more than the company assumed; perhaps the sales material is answering the wrong question first.<\/p>\n

The writing improves because the underlying material improves. You are no longer asking AI to manufacture an interesting point from a generic subject.<\/p>\n

AI writing becomes obvious through repetition<\/h2>\n

There is no single word that proves somebody used AI.<\/p>\n

Humans write \u201chowever\u201d. They use em dashes. They occasionally say something is significant. A lawyer, asset manager or technology specialist may use formal terminology because formal terminology is appropriate to the subject.<\/p>\n

What becomes noticeable is repetition across a longer piece.<\/p>\n

You may find five sentences built in almost the same way. Paragraph after paragraph introduces a claim, explains it and finishes with a neat general conclusion. Every section has a similar length. The copy repeatedly tells you that a development \u201chighlights\u201d, \u201cunderscores\u201d or \u201cdemonstrates\u201d something, even though the previous sentences already made the point.<\/p>\n

GPTZero and Pangram approach detection through broader patterns as well. Their systems analyse features of language and structure and estimate whether a passage resembles text produced by AI. They are not simply searching for a blacklist of suspicious expressions.<\/p>\n

An editor can do something similar without the software. Read four or five paragraphs together rather than inspecting one sentence at a time. We can miss the regularity when every sentence looks reasonable in isolation, but it becomes much more obvious when the article keeps moving at exactly the same pace.<\/p>\n

This is also why swapping formal words for simpler ones only gets you so far. Changing \u201cutilise\u201d to \u201cuse\u201d will improve the sentence, but it will not change an article whose paragraphs still follow the same template.<\/p>\n

Say what the company actually did<\/h2>\n

A lot of AI language is really corporate language with the volume turned up.\u00a0Marketing teams talk about \u201cimproving engagement\u201d, \u201coptimising journeys\u201d, \u201cbuilding trust\u201d and \u201cdriving growth\u201d. Those phrases can be useful shorthand inside a meeting because everybody already knows what project is being discussed. In an article, the reader does not have that background.<\/p>\n

Suppose engagement went up after the content team reduced a newsletter from eight stories to four and moved the investment commentary to the top. That is more interesting than saying the company \u201coptimised its content strategy\u201d.<\/p>\n

Perhaps a wealth manager wanted to make fees easier to understand and began publishing them before the first adviser meeting. Write that rather than saying it \u201cstrengthened transparency\u201d.<\/p>\n

If mobile users kept abandoning an application until the digital team removed a compulsory telephone-number field, the reader will learn more from the abandoned form than from a sentence about improving the customer journey.<\/p>\n

Writing this way also fixes another problem that appears frequently in AI-assisted copy: nobody seems to do anything. \u201cThe implementation of the new strategy resulted in an increase in engagement\u201d hides the people and decisions behind abstract nouns. When you know who changed the newsletter, pricing page or application form, put them back into the sentence.<\/p>\n

You do not have to remove every nominalisation from professional writing. Some terminology genuinely belongs in the sector. You simply want enough people, companies and actions in the copy that readers can see what happened.<\/p>\n

You rarely need to tell people why the previous sentence matters<\/h2>\n

One of the habits I would check first in any AI-assisted draft is how often the article stops to comment on its own argument.<\/p>\n

\u201cThis matters for marketers.\u201d<\/p>\n

\u201cThis represents a significant shift.\u201d<\/p>\n

\u201cThe implications are clear.\u201d<\/p>\n

\u201cThis creates an opportunity for brands.\u201d<\/p>\n

Any of those sentences can appear in perfectly good human writing. The problem comes when they appear repeatedly because the draft keeps announcing a conclusion that could have been explained instead.<\/p>\n

If a privacy rule limits the audience data available to advertisers, tell the reader what the team now has to change. It may need to build more campaigns around first-party data, accept less precise targeting or move spending towards channels where the company already has a relationship with the customer.<\/p>\n

A competitor cutting prices is another example. Calling the move \u201csignificant\u201d does not tell a marketing director much. If it forces the sales team to defend a premium more often during renewals, attracts a new customer segment or changes the offers needed in paid campaigns, write about that.<\/p>\n

You will often find that once the consequence appears in the paragraph, the sentence explaining that the development is important can simply go.<\/p>\n

Do not manufacture rhythm<\/h2>\n

People often advise writers to vary sentence length when they want AI-assisted copy to sound more natural. That advice is useful until it becomes another formula.<\/p>\n

You have probably seen the pattern by now: a long explanatory sentence is followed by a five-word statement intended to land with extra weight.<\/p>\n

The lesson is clear.<\/p>\n

Everything has changed.<\/p>\n

Trust is what matters.<\/p>\n

A short sentence can be effective when the argument genuinely calls for one. If the same device appears every few paragraphs, it stops sounding spontaneous and starts sounding produced.<\/p>\n

Good rhythm comes more naturally when you allow different ideas to take up different amounts of space. One example may need a longer sentence because there is an important qualification. Another point may only need ten words. A company case might deserve two paragraphs because the result is complicated, while a secondary example can sit inside one sentence.<\/p>\n

The same is true of sentence construction. If several sentences begin with the company name and then move through an action and a result, the paragraph becomes predictable. Sometimes a date belongs first because chronology explains the story. Sometimes the qualification has to come before the claim because the claim only holds under certain conditions. Sometimes the comparison is the most natural starting point.<\/p>\n

You do not have to plan that variation sentence by sentence. In fact, doing so often makes the prose feel more artificial. Concentrate on the relationship between the ideas and the syntax usually starts to vary by itself.<\/p>\n

Punctuation works much the same way. Parentheses are useful when you have a relevant aside or qualification. Colons and semicolons have perfectly ordinary jobs. Em dashes are fine too, but using them several times on every page quickly becomes noticeable because they have become such a common feature of AI-assisted prose.<\/p>\n

Be careful with sentences that sound too complete<\/h2>\n

AI is very good at producing a sentence that feels finished before you have had time to question whether the idea is accurate.<\/p>\n

\u201cMarketing is not about reach; it is about relevance.\u201d<\/p>\n

\u201cAI is not replacing creativity; it is augmenting it.\u201d<\/p>\n

\u201cClients do not want more communication. They want better communication.\u201d<\/p>\n

They work because the contrast gives the sentence a satisfying rhythm, although real marketing decisions are rarely organised so neatly.<\/p>\n

Reach and relevance can both determine whether a campaign works. AI already replaces some production tasks while helping people with others. Some clients may want fewer routine emails and much more communication when markets become volatile.<\/p>\n

Once you describe that relationship properly, the sentence becomes less polished in the slogan sense and more useful professionally.<\/p>\n

You can see something similar with groups of three. AI has a fondness for phrases such as \u201cclarity, consistency and credibility\u201d because three items create an easy cadence. Three is fine when you actually have three separate factors. It is worth checking, however, whether the third exists because the analysis requires it or because the sentence sounded unfinished with two.<\/p>\n

What GPTZero and Pangram can tell you<\/h2>\n

GPTZero is one of the better-known AI-text detectors. It analyses linguistic and statistical features across a document and estimates whether the text resembles AI-generated writing.<\/p>\n

Pangram also uses trained classifiers to distinguish patterns found in human and machine-generated material. Substack has integrated Pangram into its platform, allowing readers to scan eligible posts and notes and receive an estimate of how much of the material appears human-written or AI-assisted.<\/p>\n

That development is worth knowing about if you produce thought leadership, newsletters or executive content because detection is moving into the same environments where people actually read the work.<\/p>\n

It still does not give us a simple answer to the question \u201cWho wrote this?\u201d<\/p>\n

Imagine three marketing professionals. One writes an entire article and asks AI to shorten it. Another uses AI to organise research and then writes the published copy from scratch. A third asks for a full first draft and subsequently rewrites most of it. A detector can examine the resulting language, but it cannot reliably reconstruct the decisions behind those three workflows.<\/p>\n

False positives and false negatives are possible too. Someone who writes in a highly regular or formal style may be classified as AI even when they wrote every word themselves. Heavily edited AI copy can move in the other direction.<\/p>\n

I would therefore use a detector as another editing tool rather than treating its score as a verdict. If the same passage keeps getting flagged, read it again. You may find overly regular syntax, vague business language or three paragraphs making essentially the same argument.<\/p>\n

The detector can be wrong about authorship and still draw your attention to something worth editing.<\/p>\n

Claude’s watermark works differently<\/h2>\n

Anthropic’s approach with Claude is technically different because it does not depend entirely on judging whether the prose looks like AI.<\/p>\n

Newer Claude models can make subtle changes in token selection while generating text, leaving an invisible statistical watermark. A compatible detector can later look for that signal and estimate whether Claude was involved in producing the text.<\/p>\n

That still tells you less than people sometimes assume.<\/p>\n

Claude could have generated the first draft, rewritten something written by a person or made substantial edits to existing copy. A watermark can indicate involvement without explaining what kind of involvement took place. It does not tell you who came up with the idea, who researched the article or who decided what the conclusion should be.<\/p>\n

Anthropic also says substantial rewriting can weaken or remove the signal. In practical terms, watermark detection and tools such as GPTZero or Pangram are looking at different things: one searches for a provenance signal deliberately introduced during generation, while the others infer probable AI involvement from characteristics of the finished language.<\/p>\n

Neither can tell you whether a marketing article contains an observation worth publishing.<\/p>\n

Give AI more editing work<\/h2>\n

For many experienced marketers, AI becomes more useful once you stop asking it to be the author of everything.<\/p>\n

Give it the interview transcript and ask which themes keep appearing. Ask it to compare two structures. Get it to identify places where your conclusion goes beyond the evidence or where paragraph four simply repeats paragraph two.<\/p>\n

It is particularly good at the kind of final check that becomes difficult after you have stared at the same draft for two hours.<\/p>\n

Ask which sentences could appear unchanged in another industry. Ask whether five consecutive paragraphs follow the same pattern. Ask where nouns have replaced stronger verbs. Ask where you keep joining sentences with \u201cand\u201d, where every section finishes with a summary or where you have created a neat contrast that the evidence does not really support.<\/p>\n

Then make the decisions yourself.<\/p>\n

You know whether one client’s comment is representative or unusual. You know why management changed its view of a campaign. You may remember that the apparently successful competitor strategy failed when your company tried something similar three years ago. AI will not automatically know any of that unless you give it the information, and even then you are still the person who understands how much weight it deserves.<\/p>\n

Those choices are where a professional voice usually comes from.<\/p>\n

Sometimes starting again is quicker<\/h2>\n

Editing directly on top of an AI-generated draft has one disadvantage: the original sentence architecture tends to survive longer than you expect.<\/p>\n

You may replace the obvious vocabulary and delete a few clich\u00e9s, yet the paragraph still begins with a general statement, introduces a tidy contrast halfway through and ends with a broad lesson.<\/p>\n

For important external articles, I often think the better method is to keep the useful research and structure, close the generated prose and write the key section again from your notes.<\/p>\n

Once you stop looking at the original wording, you have to decide what the point actually was. You remember the example you found interesting, skip an explanation that felt unnecessary and spend longer on the part where you know something the model did not.<\/p>\n

The process takes a little more time than changing synonyms, but much less than endlessly trying to make a generic paragraph sound personal.<\/p>\n

Ask whether somebody can actually own the argument<\/h2>\n

Before publishing an AI-assisted article, we would use two simple checks.<\/p>\n

First, remove the company and product names from a paragraph. If a private bank, luxury retailer, software company and hotel group could all publish it without changing the underlying argument, there probably is not enough information in it yet.<\/p>\n

Then imagine somebody asking the named author about the article in a meeting. Where did that number come from? Why was this example chosen? What makes you think one thing caused another? What evidence would make you change your conclusion?<\/p>\n

A person does not need to have manually typed every word to answer those questions, but they should understand and stand behind what is being published under their name.<\/p>\n

That is also where AI becomes genuinely useful at work. There is little reason to spend an hour manually sorting 20 pages of interview notes if a tool can do it in a minute, and nobody wins a prize for personally generating twelve variations of a headline. Use the saved time to speak to the sales team, check the figures, read the original report or find the example that makes the argument more precise.<\/p>\n

Once the article contains enough of that work, you usually do not have to try very hard to make it sound human. It sounds different because it contains somebody’s knowledge, somebody’s judgement and enough specific detail that another marketer could not simply change the company name and publish the same piece tomorrow.<\/p>\n

 <\/p>","protected":false},"excerpt":{"rendered":"

Learn the art of leveraging AI without losing the human touch. <\/p>","protected":false},"author":5,"featured_media":13916,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[117,77],"tags":[],"class_list":["post-13914","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-content-marketing","category-insights"],"yoast_head":"\nHow To Use AI At Work Without Sounding Like AI - Exporis<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.exporis.ch\/it\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\" \/>\n<meta property=\"og:locale\" content=\"it_IT\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"How To Use AI At Work Without Sounding Like AI - Exporis\" \/>\n<meta property=\"og:description\" content=\"Learn the art of leveraging AI without losing the human touch.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.exporis.ch\/it\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\" \/>\n<meta property=\"og:site_name\" content=\"Exporis\" \/>\n<meta property=\"article:published_time\" content=\"2026-08-18T05:12:05+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-08-18T06:41:39+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1920\" \/>\n\t<meta property=\"og:image:height\" content=\"1080\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Cathy\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Scritto da\" \/>\n\t<meta name=\"twitter:data1\" content=\"Cathy\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tempo di lettura stimato\" \/>\n\t<meta name=\"twitter:data2\" content=\"13 minuti\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\"},\"author\":{\"name\":\"Cathy\",\"@id\":\"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50\"},\"headline\":\"How To Use AI At Work Without Sounding Like AI\",\"datePublished\":\"2026-08-18T05:12:05+00:00\",\"dateModified\":\"2026-08-18T06:41:39+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\"},\"wordCount\":2810,\"image\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg\",\"articleSection\":[\"Content Marketing\",\"Insights\"],\"inLanguage\":\"it-IT\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\",\"url\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\",\"name\":\"How To Use AI At Work Without Sounding Like AI - Exporis\",\"isPartOf\":{\"@id\":\"https:\/\/www.exporis.ch\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg\",\"datePublished\":\"2026-08-18T05:12:05+00:00\",\"dateModified\":\"2026-08-18T06:41:39+00:00\",\"author\":{\"@id\":\"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50\"},\"breadcrumb\":{\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#breadcrumb\"},\"inLanguage\":\"it-IT\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"it-IT\",\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage\",\"url\":\"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg\",\"contentUrl\":\"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg\",\"width\":1920,\"height\":1080},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/www.exporis.ch\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How To Use AI At Work Without Sounding Like AI\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/www.exporis.ch\/#website\",\"url\":\"https:\/\/www.exporis.ch\/\",\"name\":\"Exporis\",\"description\":\"Stay Ahead Of The Game\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/www.exporis.ch\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"it-IT\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50\",\"name\":\"Cathy\",\"url\":\"https:\/\/www.exporis.ch\/it\/author\/cathy\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"How To Use AI At Work Without Sounding Like AI - Exporis","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.exporis.ch\/it\/how-to-use-ai-at-work-without-sounding-like-ai-2\/","og_locale":"it_IT","og_type":"article","og_title":"How To Use AI At Work Without Sounding Like AI - Exporis","og_description":"Learn the art of leveraging AI without losing the human touch.","og_url":"https:\/\/www.exporis.ch\/it\/how-to-use-ai-at-work-without-sounding-like-ai-2\/","og_site_name":"Exporis","article_published_time":"2026-08-18T05:12:05+00:00","article_modified_time":"2026-08-18T06:41:39+00:00","og_image":[{"width":1920,"height":1080,"url":"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg","type":"image\/jpeg"}],"author":"Cathy","twitter_card":"summary_large_image","twitter_misc":{"Scritto da":"Cathy","Tempo di lettura stimato":"13 minuti"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#article","isPartOf":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/"},"author":{"name":"Cathy","@id":"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50"},"headline":"How To Use AI At Work Without Sounding Like AI","datePublished":"2026-08-18T05:12:05+00:00","dateModified":"2026-08-18T06:41:39+00:00","mainEntityOfPage":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/"},"wordCount":2810,"image":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage"},"thumbnailUrl":"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg","articleSection":["Content Marketing","Insights"],"inLanguage":"it-IT"},{"@type":"WebPage","@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/","url":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/","name":"How To Use AI At Work Without Sounding Like AI - Exporis","isPartOf":{"@id":"https:\/\/www.exporis.ch\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage"},"image":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage"},"thumbnailUrl":"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg","datePublished":"2026-08-18T05:12:05+00:00","dateModified":"2026-08-18T06:41:39+00:00","author":{"@id":"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50"},"breadcrumb":{"@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#breadcrumb"},"inLanguage":"it-IT","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/"]}]},{"@type":"ImageObject","inLanguage":"it-IT","@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#primaryimage","url":"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg","contentUrl":"https:\/\/www.exporis.ch\/wp-content\/uploads\/2026\/08\/ai-work.jpg","width":1920,"height":1080},{"@type":"BreadcrumbList","@id":"https:\/\/www.exporis.ch\/how-to-use-ai-at-work-without-sounding-like-ai-2\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.exporis.ch\/"},{"@type":"ListItem","position":2,"name":"How To Use AI At Work Without Sounding Like AI"}]},{"@type":"WebSite","@id":"https:\/\/www.exporis.ch\/#website","url":"https:\/\/www.exporis.ch\/","name":"Esportazione","description":"Rimani sempre un passo avanti","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.exporis.ch\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"it-IT"},{"@type":"Person","@id":"https:\/\/www.exporis.ch\/#\/schema\/person\/5730d61de1e8de711efd0476234e1e50","name":"Cathy","url":"https:\/\/www.exporis.ch\/it\/author\/cathy\/"}]}},"_links":{"self":[{"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/posts\/13914","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/users\/5"}],"replies":[{"embeddable":true,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/comments?post=13914"}],"version-history":[{"count":1,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/posts\/13914\/revisions"}],"predecessor-version":[{"id":13917,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/posts\/13914\/revisions\/13917"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/media\/13916"}],"wp:attachment":[{"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/media?parent=13914"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/categories?post=13914"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.exporis.ch\/it\/wp-json\/wp\/v2\/tags?post=13914"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}