 |
|
|
|||||
Have you ever read an article or social post and thought, This is terrible! I bet it was written by AI!? Most people know bad AI writing when they see it. But unless youre a closeted copy editor, its surprisingly hard to put your finger on exactly why AI writing sucks. Now, Wikipedias editor team has just released what amounts to a master class in the clichés, strange tropes, obsequious tones of voice, and other assorted oddities of AI-generated prose. Its a list called Signs of AI Writing, and its a fantastic resource for people who want to get better at spotting AI writingor who want to disguise their own. Add your own slop As one of the internets most trusted sources of information, Wikipedia is uniquely exposed to the risks of LLM-generated content. Large language models love to pontificate on random topics, even when they have very little actual knowledge. Wikipedia covers many of these random topics, from the ash content of Morbier cheese to the gorey details of Justin Biebers love life. Wikipedia famously crowdsources its information through a network of volunteer contributors and editors. This combination of crowdsourced data and highly specific, niche topics is a recipe for the misuse of AI. Theres also an increasingly potent financial incentive for people to pollute Wikipedia with AI slop. As search engines like Google laser in on EEATa tortured acronym that describes the authoritativeness of a brandhaving a Wikipedia page is becoming more valuable to brands as a metric for their legitimacy. Youre not supposed to create or edit your own Wikipedia page, but many brands do. And one of the easiest ways to hide this off-label tinkering is to drown ones nefarious edits in a sea of seemingly unrelated updates and contributions to esoteric Wikipedia pages. AI can spin these up at scale. Everything is fascinating Because of the risks that AI-generated content poses to the site, Wikipedias editors have gotten incredibly good at recognizing AI writing. Their Signs of AI Writing document distills this knowledge into an easy-to-follow guide. Wikipedias list is useful and unique largely because its so specific. Many other rubrics for recognizing AI writing offer broad, generic advice or focus on detection hacks that are easy to bypass. Researchers recently realized, for example, that LLMs tend to overuse the em dasha wonderful and remarkably versatile punctuation mark that I happen to absolutely love. As I recently discussed with Slate, for a brief moment, the presence of an em dash in an article was a good way to detect AI writing. Quickly, though, AI content generators caught on and started to avoid the punctuation mark. Simple hacks for detecting AI writing have a limited shelf life. The arms race of AI content creation and AI content detection means these methods are quickly rendered useless as soon as theyre made public. Wikipedias guidelines go much deeper. Rather than focusing on quick detection hacks, they dig into the more fundamental patterns present in bad AI contentthe writing conventions and literary tropes that LLMs consistently overuse. Wikipedias editors point out, for example, that LLMs place undue emphasis on symbolism and importance. Everything LLMs write stands as a symbol of something, or carries enhanced significance. Natural locations are always captivating, all animals are majestic and everything is diverse and fascinating. Wikipedias editors also note that LLMs tend to overuse transition words and phrases like in summary or overall. Often these show up as negative parallelisms. For example, LLMs love to summarize things theyre already written with tropes like: Its not only but also A restaurant might be described as not only a great place for Italian food, but also a shining example of local entrepreneurship. Every concluding paragraph starts with In conclusion or In summary. The editors also point out that AI writing often overuses the Rule of Three–a handy literary trick that capitalizes on the fact that humans brains love groups of three things. A person might be creative, smart and funny according to ChatGPT, or a company could be innovative, rule-breaking and impactful. Good writing gone bad Interestingly, Wikipedias editors acknowledge that many of these conventions would be considered good writing if they came from a human. Its not that LLMs are inherently bad at writingits just that they write in predictable ways that make their output feel formulaic and robotic. The editors also note that LLMs polished writing style and tendency to follow conventions often serves to obscure their lack of actual knowledge about a topic. By following conventions like the Rule of Three, LLMs make their superficial explanations appear more comprehensive. As readers, we often mistake good form for good contentif an LLM writes with perfect grammar and its content flows beautifully, we might not realize that its not actually saying anything useful or substantive. Beyond these stylistic issues, Wikipedias list goes into extreme detail about technical specifics of AI writingthe ways LLMs consistently format text, use headings, handle punctuation (like curly quotation marks), and sprinkle their content with bolded words and emojs. Spot it (or make it) The guidelines are useful for anyone who edits Wikipedia. But theyre also relevant for anyone who wants to get better at recognizing AI writingor who wants to create their own AI content that doesnt sound machine-generated. If youre reading an article or social media post that feels a bit off and youre curious whether it might be AI-written, Wikipedias guidelines provide a fantastic checklist for validating your suspicions. Compare the suspect writing with Wikipedias list. Do you see the Rule of Three appear a bit too consistently? Are there too many transition words? Does it sound too effusive? Although the editors stress that humans are perfectly capable of generating bland and formulaic writing without an AIs help, spotting these patterns in a piece of writing can lend credence to the idea that it was written by a machine. And if you use LLMs to create content for your businessor even for personal emails or social postsWikipedias list can help you tweak it so its genuinely readable and doesnt sound quite so robotic. As a human editor, you can manually scan the output of ChatGPT, Claude or Gemini for the patterns Wikipedia identifies, and inject your own human touch when the chatbots start sounding a bit too AI. Theres an easier approach, too. Ive found that pasting Wikipedias entire Signs of Writing list into a chatbot as part of your prompt yields noticeably better writing than LLMs produce alone. Spinning up a social post for your bands first mall gig, or generating the landing page copy for your crochet business Square page? Prompt ChatGPT or Claude as you normally would, but tell the chatbot to avoid the items on this list. Then, paste in the full contents of Wikipedias Signs page. Your LLM-generated writing will feel markedly better, with very little effort. Make sure to use your powers for good! With their specificity, focus on stylistic rather than technical patterns, and attention to subtle details of AI writing (see, Rule of Three!), Wikipedias list is a fantastic tool for anyone who wants to spot lazy AI writingor make their own AI content feel a bit less lazy and generic.
Category:
E-Commerce
Its not just the tech industry that is being battered by mass layoffs this year. Grocery store giant The Kroger Co. (NYSE: KR) is cutting nearly 1,000 jobs from its corporate workforce. Heres why, and how the companys stock is reacting. Whats happened? Yesterday, Kroger interim CEO Ron Sargent said that the grocery chain would lay off hundreds of corporate workers, according to a memo seen by Fast Company. The layoffs will total fewer than 1,000 employees. Kroger currently employs around 409,000 workers, the majority of whom work in its 2,700 grocery stores, which include Kroger, Food4Less, CityMarket, and more. In the memo, Sargent revealed that, “In the past few months, we have all looked for ways to simplify the organization, shift resources closer to our customers, and focus on work that creates the most value. However, the job cuts will not affect employees in the companys stores, distribution centers, or manufacturing facilities. The memo went on to say that the savings from the corporate layoffs would be reinvested in the company and used to help fund new locations, create store-level jobs, and offer price reductions to customers. The layoffs and memo were reported earlier by Reuters. A Kroger spokesperson confirmed the job cuts when contacted by Fast Company. Layoffs follow store closures and failed merger The newly announced layoffs mark another low point for Kroger over the past 12 months. Within that timeframe, the company has incurred significant setbacks. The most dramatic of those is the failed merger between Kroger and Albertsons, which was valued at $25 billion. The merger would have seen the two grocery store chain giants join ranks, creating a new supermarket juggernaut. This would have allowed the newly formed company to compete against grocery offerings from arch-rivals Walmart and Amazon. However, in December, a federal judge blocked the merger on anti-competition fears. The merger was later abandoned entirely and has led to legal proceedings between the companies. Kroger has also announced this year the closure of 60 of its stores, which are expected to shutter by the end of 2026. Kroger said that the closures would provide it a modest financial benefit. Kroger investors shrug off the job cuts Despite the devastating effect that these layoffs will have on the impacted workers, the news seems to have had no impact on Krogers stock price. As of the time of this writing, in premarket trading, KR shares are currently trading flat at $69 apiece. Thats just one cent higher than their closing price of $68.99 yesterday. Year-to-date, Kroger shares are up over 12.8%. And over the last 12 months, KR shares have climbed more than 30%. This story has been updated with Kroger’s response to our inquiry.
Category:
E-Commerce
Most enterprises treat AI implementation as a procurement problem. They evaluate vendors, negotiate contracts, and deploy solutions. But this transactional approach misses a fundamental truth: successful AI implementation isn’t just about buying technologyit’s about orchestrating an ecosystem. The companies winning with AI understand that implementation requires a web of relationships extending far beyond traditional vendor partnerships. They are building networks that include universities, regulatory bodies, ethicists, suppliers, and even customers. They recognize that in an environment in which AI capabilities evolve monthly, isolated implementation is a recipe for obsolescence. This article draws on insights from our forthcoming book, Reimagining Government (Faisal Hoque, Erik Nelson, Tom Davenport, Paul Scade, et al.) identifying the key components you will need to reconcile to successfully orchestrate a comprehensive AI partner ecosystem. The Expanding Universe of AI Partners When enterprise leaders think about AI partnerships, they typically start and stop with technology vendors. This narrow view blinds them to the full spectrum of relationships that determine success or failure in AI implementation. Academic institutions offer capabilities that money alone can’t buy. Universities are where breakthrough AI research happens, often years before commercial availability. Building relationships with labs, research centers, and individuals academics can provide access to cutting-edge research, specialized expertise, and talent pipelines that vendors can’t replicate. Government agencies are partners, not just regulators. Forward-thinking companies will work with agencies to shape AI standards, participate in regulatory sandboxes where they can test implementations and receive guidance, and collaborate on public-private initiatives that define industry practices. Ethics and oversight partners are becoming essential as AI stakes rise. Third-party ethicists provide a layer of credibility that internal roles cant match. Audit firms specializing in AI bias detection offer independent validation. Compliance specialists navigate the emerging patchwork of AI regulations. These partners don’t just reduce riskthey become competitive differentiators when customers demand proof of responsible AI use. Consultants and implementers bridge the gap between AI potential and operational reality. They build custom tools that integrate AI into existing workflows, train teams on new capabilities, and manage the organizational change that AI demands. The best ones will transfer knowledge while implementing systems, building internal capabilities that will endure after they leave. Supply chain partners determine whether AI creates value or chaos. When your AI-optimized inventory system hands off to a supplier’s manual processes, many of the benefits evaporate. Enterprises should look to coordinate AI decisions across their supply networks, encouraging shared model adoption, and ensuring that AI-to-AI handoffs work seamlessly. Customers are perhaps the most overlooked partners in AI implementation. They’re not just users but cocreators, providing the feedback that shapes AI development, the data that improves models, and the trust that makes implementation possible. Strategic Imperatives for Partnership Design Building an AI ecosystem that creates value involves more than just accumulating partners. Relationships and networks need to be designed to amplify capability while maintaining flexibility. Enterprises should focus on: Interoperability by design. Using proprietary models can lead to the creation of silos within enterprise networks. Selecting open-weight models helps ensure transparency and compatibility among partners. Alignment across the value chain. A pharmaceutical company implementing AI for drug discovery must ensure that contract research organizations, clinical trial partners, and regulatory consultants all work with compatible systems and standards. This doesn’t mean that all partners must use identical tools, but it does mean establishing common data formats, shared evaluation metrics, and aligned security protocols. Risk distribution. AI failures can cascade through networks. Smart partnership agreements distribute both opportunity and liability, ensuring that no single partner bears catastrophic risk while maintaining incentives for responsible development. This includes technical risks (system failures), ethical risks (bias, privacy violations), and business risks (market rejection, regulatory penalties). Translation layers. When government agencies partner with commercial vendors, they often use specialized contractors who serve as a critical middle layer, translating the generally applicable technology to meet agency-specific requirements. This middle layer adapts cloud-native solutions for secure environments, restructures Silicon Valley business models for public sector procurement cycles, and bridges cultural gaps between tech innovation and public service. Private enterprises can adopt this model as well, using specialized partners to translate general-purpose AI products for their specific industry needs. These translation partners can package the technical adaptation skills, business model alignment know-how, and cultural bridging that turns raw AI capability into operational value. Critical Partnership Challenges Three challenges consistently derail AI partnership ecosystems. The IP question can become extremely complex in multiparty AI development. When your data trains a vendor’s model that’s customized by a consultant and integrated by a systems implementer, who owns what? Imagine that a financial services firm discovers their AI vendor is using patterns learned from their fraud detection system to improve products sold to competitors. This might be permissible under the vendors standard contract, so it is important to think ahead to ensure that explicit boundaries are drawn between vendor improvements and innovations rooted in the clients operations and data. Lock-in risks extend beyond technology to psychology. Technical lock-in is a familiar problem: specific vendor systems can become so deeply integrated that switching becomes prohibitively expensive or onerous. But psychological lock-in is just as dangerous. Teams can become comfortable with familiar interfaces, develop relationships with vendor personnel, and resist exploring alternatives even when superior options emerge. Coordination complexity multiplies with each partner. When an AI system requires inputs from five partners, processes from three more, and delivers outputs to 10 others, coordination becomes a full-time job. Version mismatches, update conflicts, and finger-pointing when problems arise can paralyze initiatives. Building Your Partnership Strategy Creating an effective AI ecosystem requires a systematic approach, not just building a sequence of ad hoc relationships. Map your ecosystem needs across every dimension. Where are your technology gaps? Which expertise is missing? What ethical oversight do you need? How will implementation happen? Don’t just list vendorsmap the full spectrum of partnerships required for successful AI implementation. Include the nonobvious: the anthropologist who understands how your customers actually behave, the regulator who’ll evaluate your system, the supplier whose cooperation determines success. Design for flexibility. AI capabilities change monthly. Build partnerships that can evolve with them, with regular review cycles, clear performance metrics, and graceful exit provisions. Avoid agreements that lock you into specific technologies or approaches. The perfect partner for today’s needs may be obsolete tomorrow. Create governance structures that acknowledge the complexity of AI partnership networks. Establish steering committees with senior representation from key partners. Define escalation paths before problems arise. Create shared metrics that reflect interconnected outcomeswhen success requires five partners working together, individual KPIs create dysfunction. Plan exits from day one. As we emphasize in our recent book Transcend, knowing how partnerships end is as important as knowing how they begin. Define termination triggers, data ownership post-partnership, and transition procedures. The best partnerships are those either party can leave without destroying value. The AI revolution will not be won by technological advances alone. The strength of an enterprises ecosystem will play a key role in separating the winners from the losers. Companies that can see past traditional vendor relationships to orchestrate comprehensive partnership networks will transform AI from an implementation challenge into a sustainable competitive advantage.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||