As stated in the whatsapp group, I posted my first ChatGPT based review and it got published yesterday (03 May 2023). You can find it here:
https://www.worldheritagesite.org/tentative/id/6546#userreview_19375

I have to admit that I had been wanting to write this review for a while (see creation date back in January), but never got around to it. I would call it writer's block. ChatGPT solved it, as it essentially helped a ton on the boilerplate text. I wanted to give some context on the Umayyad's and Hisham and the site itself and it took me too long to research as I am not well versed in the history of early Islam. ChatGPT solved that. It also helped me write like a pro/native speaker. It did not, though, define how I wanted to tell my review, nor provide my impressions. To me, it served more as my personal research assistant and text doctor:

1) Ask the prompt to give me some text on the Umayyads and Hisham. Focus the prompt on details I find important.
2) Rewrite it and clean up the flowery language. Add my personal observations and key messages.
3) Ask the prompt to clean this up.
4) Finalize. Prompt will make it less personal and too flowery again.

Some observations:
1) It's hard to make the prompt go negative or personal. It generally puts a positive, generalist spin on topics. So, you can't just use it as is, at least for a personal experience.
2) I am not sure how much time I saved, but the presentation of the context information is way better and better summarised than it would have been, if I had tinkered on my own.
3) My writer's block primarily stemmed from too high quality expectations. Ever since using ChatGPT, my output has increased significantly. See my upcoming reviews. So this seems addressed.
4) I still have to tinker a bit how to make it work best for me. But if allowed to continue, I will make use of Generative AI going forward.

Questions to discuss:
1) What's your impression re my review(s)?
2) Do you use or want to use ChatGPT in your own reviews?
3) When do we consider a review to be authentic and personal and okay to publish?

Looking forward.

Thanks for that Nan! Very interesting. I have now tried ChatGPT asking it the question "Please produce a review of a a visit to XXXXX" I have also done so multiple times
a. With exactly the same question
b. With slightly varied questions for the same place

I can't say that the result has met the "expectations" which have been raised by all the media hype about AI as exemplified by ChatGPT - indeed I was rather disappointed with them and wouldn't want to use any of them . or even any bits of any of them!!!
a. OK - it comes up with reasonably coherent short essay in acceptable English. Unfortunately it reads like something written by a US High schooler on "happy pills". Everything is "wonderful"... "the staff at XXXX are very freiendly and welcoming", XXXXX is "a fascinating and historic site", "The site was well-maintained, and there were plenty of information boards", "One of the highlights of a visit to XXXXX is the stunning views from the top", I presume the Bot has access to a knowledge base which includes critical "reviews" but it doesn't access these unless pushed to do so
b. I tried 5 versions with the same question - all were much of a muchness - the same facts and opinions regurgitated in a slightly different order.
c. I tried "please produce a critical review" and "please produce a mixed review". It did this but it wasn't clear where it found its alternative views from - perhaps it found some Trip Advisor reviews which (in this case) criticised the maintenance of the site? (I also asked it to "produce a bad review" - but it refused on "ethical" grounds - so I had to change the word "bad" to "critical"!!!)

Clearly one can't expect the Bot to replicate personal experiences - but, once one has fed these in for oneself - what is left for it to do? Perhaps those for whom English is not a first language will feel more confident in producing a review with it providing a "starter" for them?? I personally would prefer to read their "broken" English or an auto translate by Google!!

It also provides a bit of background context to "surround" the self generated personal experiences. Again, I personally would rather have a link to a proper article or to Wiki!! If everyone uses it we will get the same "facts" produced again and again!

It also actually included some "errors" about the visit experience - perhaps the knowledge base was out of date. Anything produced by it is going to need a lot of checking.

Each reviewer produces their reviews for their own reasons. If the purpose is to provide useful experiences and information for others (rather than as a "record" for themseleves) I don't see that the Bot adds anything unless it gives someone the confidence to produce a review of their experiences which they otherwise would not have done.

Fo me this is where the value of it has been thus far. It produces text that gets me past a certain block, in much the same way that my editor or press colleagues have done in my job.

Also the block being the result of my own high expectations is precisly my issue, that is not to say my reviews are of high quality (reading them back they are always riddled with errors) just that I sometimes get stuck compared to what I set out to acheive.

As I said in the WhatsApp group I rather liked the review, I felt the first section was a little fact heavy and felt a touch less human (but only a small critique as I was specifically looking for AI clues). The OUV section read as much more natural.

As you said I think it works well as an assiatant or tool for getting past certain blocks, but it does lose your specific voice as an author.

I shared your review with my wife last night, and didn't tell here about the ChatGPT aspect. She is much more erudite than I am and didn't twig that AI played a role in it at all.

I'm stuck a little on a review at the moment, perhaps I will put it through ChatGPT to see what polish it gives it, though the focus is raher negative so maybe it will come up against the limits Solivagant has also found.

Two things up front:
* 2nd ChatGPT review is out.
* I will do a review and document my steps in the process.


Doesn't work that way. You need to give it the bullets and structure and proper directions. Think of the tool more as a personal assistant who carries out laborious tasks like researching Umayyad history.


Again, imagine you had an assistant, smart but now experienced. You need to tell them what they are supposed to do.


Most reviewers on this site, me and Els included, are not native speakers. Having a tool that pimps my English to acceptable English at short notice is quite a feature. Feedback from Els on first review was that it doesn't sound like a non native speaker. Apparently, my previous posts did ;)


I noticed this too. There is a strong "positive bias" and you have to work hard to undo it. But... via prompt you can, e.g. tell it to take the view that Venice is an overrated site. And I mostly had to delete the conclusions as they tried to have a happy ending.


These are faked. Experiences you have to provide yourself. Don't use a bot to generate it. Use the bot afterwards to phrase it nicer.


References and sourcing are not the strong point of generative AI. Essentially, it generates the next plausible word and does not represent reality. I think this will change in the coming months/years.


Auto translate by google is pretty bad. As stated by both Ian and myself, the expectation to write sth sensible can prevent reviews being written at all. Generally, while you personally may prefer an authentic review, I would argue the overall internet does not.


I am not as much a fan of links as I used to. Essentially, you should not need to read a lengthy wiki page to know who Hisham was when you read a review of Hishams palace.

In addition, I generally read what previous reviews exist and keep redundant information short. Hisham, e.g., was a first review for the site, so giving some facts makes sense. My Siza's Ensemble review meanwhile is heavily focused on my experience. It does not provide context.


I think it creates a higher quality review in less time in a language I am not a native speaker of. And that's what it adds to me.


Fact heavy came from the review being the first. And then I wanted to provide a little background information. I repeatedly told ChatGPT to shorten (50 words or less), though ;) Looking at the review section, I would probably have only done two less fact-heavy paragraphs: History of Umayyads and Hisham (1). Key elements on the site (2). Then my visit (as is).


I think my voice is still there. It's in the opinion and logistics section :) Those were only cleaned up, but not written by ChatGPT.


I found the same limits (happy endings). But that's not how you should use it. But in any case, give the future a try :)

2nd review is:
https://www.worldheritagesite.org/tentative/id/6224#userreview_19694

@Paul:
* The whole OUV section is my opinion, but essentially written by ChatGPT.
* Apart from the public transport connections no facts are included. It's an opinion piece.

Side note: Wrote on my visit experience at early synagogues. But process was more complex as I wanted to get a certain opinion across. Interestingly chat gpt answered my lingering question why these early synagogues date from late antiquity when the jewish faith is hundreds of years older.

The difference I think lies in the vocabulary used. As non-native speakers, we are more limited in the number of English words that pop up in our minds. Native speakers, especially the more eloquent ones such as our reviewer Squiffy, can come up with a variety of words that exactly bring across what he means/feels.


So funny!


You shared your script with me and that was enlightening. You asked it questions like "write from a more personal perspective'' and "please also question if such a recent structure should be added to the world heritage list". It seemed like a lot of work though, working with a junior assistant.

As long as the review still reflects your own opinion and experience while visiting the site, I have no objections. Personally, I like doing the basic research myself as well, that way you can also better verify the quality of the sources used. Sometimes a little-known scientific paper can shine a totally different light on a WHS than Wikipedia does.

@nfmungard
"Bing" is a search engine I don't normally use but it keeps trying to get me to do so on my Windows PC (Microsoft never gives up!!) - yesterday it offered me "Introducing the new Bing. Your AI-powered copilot for the web". I tried out the same sorts of experiments to produce WHS reviews as I described above for ChatGBT. It came up with similar results - it had to be directed to assess any site as less than ausgezeichnet but would do so if pushed - though on somewhat arbitrary factors - it decided for instance to suggest that a potential downside of the site I had chosen might be that it could be considered too expensive at £8 for what was on offer! Where does it get these value judgements from??? Why that one? I felt that its info on the detail of the site was "better" than that from ChatGBT - BUT it did make one very big mistake by suggesting that something was "on show" at the site when in fact it can only be seen at the National Museum. It was certainly found at the site but, I am not sure it has ever been on show there. I can't find where the AI got this from.

My questions revolve around - is there anything speical about ChatGBT? Is one supplier's artificial intelligence any more "intelligent" than another's? It is clear that, like any tool, one has to get to know it and its idiosyncracies to get the best from it. Years of using "simple" Google has taught one the "searching tricks" to get better results (I can't try "Bard" as there is a wait list) and, no doubt one can "learn" how to get the best from AI too

@solivagant maybe easier to use an example.

I am currently thinking about writing a Civita de Bagnoregio review. Key bit is that I am not sure what gap it fills seeing fortified hilltop cities in Italy are plentiful, including existing WHS (Pienza, Siena, San Gimignano). It may feature a stunning panaroma. But apart from that? So I asked ChatGPT.

1) describe civita de bagnioregio
- To get an idea and to tell the machine what I am interested in.

2) what other medieval hilltop cities are there in italy?
- i know plenty myself. let's see what the machine comes up with... visited all five and roughly the same list i would have come up with.

3) anything in the val d'orcia?
- i remember pienza as a hilltop town. lets see what is added.

4) what sets bagnoregio apart from other hilltop cities?
- i gave the system my context. let's see what it comes up with as argument.

5) bagnoregio is trying to be a world heritage site. what gap would it fill if added to the world heritage list? please compare to already inscribed hilltop towns in italy.
- gap is preservation. not really convinced.

6) the preservation need is not really a good argument for filling a gap. rather the opposite. is there any cultural or historical argument where bagnoregio fills a gap compared to already inscribed italian world heritage sites?
- pushing him to make a different argument.
- system is going for etruscan.

7) what concrete etruscan heritage remain in bagnoregio? the city looks medieval in nature. not etruscan.
- it gets defensive re etruscan. i am not convinced.

8) are there any other hilltop cities in italy that have etruscan roots?
- now it lists two other tentative sites (plus others): volterra and orvieto.

9) if you compare bagnoregio to volterra and orviete, which one has more etruscan heritage?
- it admits orviete and volterra would be stronger sites re etruscan heritage.

10) seeing both orvieto and volterra have better etruscan remains, bagnoregio doesnt really fill that gap. anything else that would be justification for inscribing it on the world heritage list?
- it makes a more sensible, but weaker argument. research is done.

note: i was able to do some rather challenging research in 10min.

Continued.

A) how can you get to bagnoregio?
- cross checking if i have the connections right. i do.

B) clean this up (repeatedly)

C) write a paragraph on civita de bagnoregio. mention the view from the viewpoint, the og we enjoyed due to arriving in the morning, and the climb on the walkway.
- gives a generic summary i can use as starting point. it was actually stunning with the fog.

D) add a paragraph on the tourists arriving when we left. add a comment on my travel companion being afraid of heights and having a hard time walking up to the city.

E) combine the two paragraphs into one text.

F) add the fog. be a bit less positive.

threw everything away and rewrote it personally.

G) clean this up (repeatedly)
-done.

Summary. It took me 1h to write a fairly lengthy review in rather good English. That to me is the strength of the tool. You will have to wait for the result a bit ;)

Update. You have been privvy to a large batch of ChatGPT reviews by now. Would be curious to know what you think?

I have changed my method, though, as I found ChatGPT too often eliminated my own voice when asked to clean up a text portion. What I do now:

0) Put in template and stuff I want to say.
1) ChatGPT - Boilerplate text like an intro paragraph or a description on the key elements.
2) Nan - Write my own stuff.
3) Use DeepL write to make the language nicer.

Lesson learnt:
* ChatGPT helps doing basic research and getting past the initial writers block.
* Deepl Write cleans up my English and makes it nicer. It does not, though, eliminate my voice. My Sarazm review should be a showcase for the adjusted approach.
* Most gains are on quality (I write better English, I use more specialist terms, I can give a generic introduction summing up the history).
* Gains are less so on speed.

Questions:
What are your thoughts on my last ChatGPT supported reviews? All reviews shown on my profile page were supported by ChatGPT.
https://www.worldheritagesite.org/community/nan

Did you write a review yourself? What are your experiences?

I am beginning to wonder what added value ChatGPT provides if even the "English" comes via DeepL!
The more I use ChatGBT for "facts" the less impressed I am. I have a recent conversation with "it" regarding the Dominican Republic's T List following the comment about Zoe's "rant" assessment! - full "conversation" attached below.
Its initial attempts are quite "plausible" until one begins to probe a little. I am absolutely amazed that it can't cite where its "facts" have been picked up from!! I would have thought that this alone would prevent its use by students to fast track essays without any work - in "my day" one had to give sources for ALL "facts"!!

User
Please write an assessment of the Dominican Republic's UNESCO World Heritage Tentative list
ChatGPT
Title: Assessment of the Dominican Republic's UNESCO World Heritage Tentative List
Introduction:
The Dominican Republic, located in the Caribbean region, boasts a rich cultural and natural heritage that is recognized globally. In its pursuit of preserving and showcasing its exceptional sites, the country has compiled a Tentative List of potential UNESCO World Heritage sites. This assessment aims to evaluate the strengths and weaknesses of the Dominican Republic's tentative list, highlighting notable sites and offering recommendations for future considerations.

Strengths:
Outstanding Natural Landscapes:
The Dominican Republic's tentative list includes several breathtaking natural landscapes, such as the Sierra de Bahoruco National Park, Jaragua National Park, and the Cordillera Central Biosphere Reserve. These areas exhibit exceptional biodiversity, unique geological formations, and represent valuable ecosystems worth protecting.
Cultural and Historical Significance:
The Dominican Republic's tentative list showcases its rich cultural and historical heritage. Sites like the Colonial City of Santo Domingo, the first European city in the Americas, present a remarkable blend of architectural styles and historical significance. The archaeological sites of La Isabela and the San Francisco Monastery are also noteworthy, providing insights into the country's colonial past.
Afro-Caribbean Culture and Heritage:
The inclusion of the Afro-Caribbean Culture and Heritage in the tentative list demonstrates the Dominican Republic's commitment to preserving the intangible aspects of its cultural heritage. This recognition acknowledges the significant contributions of Afro-Caribbean communities to the nation's social fabric, music, dance, and religious practices.

Weaknesses:
Underrepresented Indigenous Heritage:
The tentative list could benefit from a stronger representation of the country's indigenous heritage. The Dominican Republic is home to ancient Taíno settlements and ceremonial sites that hold immense historical and cultural value. Ensuring their inclusion would enrich the narrative and provide a comprehensive portrayal of the nation's heritage.
Lack of Site Development Plans:
While the tentative list features remarkable sites, it is crucial to have detailed development plans for each site. These plans should encompass sustainable tourism strategies, conservation measures, and community involvement to ensure long-term preservation and responsible management. Clear guidelines and sustainable practices are necessary to safeguard the integrity of these potential World Heritage sites.
Limited International Promotion:
The Dominican Republic's tentative list could benefit from enhanced international promotion and awareness. Increased efforts to communicate the significance and unique features of the listed sites would attract global attention, fostering tourism and international collaboration in conservation efforts.

Recommendations:
Strengthen Indigenous Representation:
The Dominican Republic should collaborate with indigenous communities to identify and include significant indigenous heritage sites in its tentative list. Recognizing and preserving these sites will promote cultural diversity and provide a comprehensive understanding of the country's history.
Develop Site Management Plans:
The authorities should prioritize the development of comprehensive site management plans for each potential World Heritage site. These plans should focus on sustainable tourism, conservation strategies, risk mitigation, and the engagement of local communities. Ensuring effective management practices will preserve the sites' authenticity and foster their long-term sustainability.
Increase International Promotion:
To raise global awareness of the Dominican Republic's potential World Heritage sites, the country should undertake targeted promotional campaigns. These efforts could include online platforms, international conferences, cultural exchanges, and collaborations with UNESCO and other relevant organizations. By showcasing the unique cultural and natural heritage of the country, the Dominican Republic can attract more visitors and encourage international support for preservation initiatives.

Conclusion:
The Dominican Republic's tentative list for UNESCO World Heritage sites demonstrates its commitment to conserving and promoting its exceptional cultural and natural heritage. By addressing the identified weaknesses and implementing the recommendations outlined above, the country can strengthen its case for inscription, ensuring the long-term preservation and recognition of its remarkable sites.

User
Adjust to be more of a rant
ChatGPT
Title: Critical Assessment of the Dominican Republic's UNESCO World Heritage Tentative List
Introduction:
The Dominican Republic's UNESCO World Heritage Tentative List is a disappointing compilation that fails to adequately represent and showcase the country's true cultural and natural treasures. This assessment aims to shed light on the glaring weaknesses and missed opportunities within the list, highlighting the urgent need for improvement and a more comprehensive approach to heritage preservation.
Weaknesses:
Neglect of Indigenous Heritage:
It is deeply disheartening to see the Dominican Republic's tentative list largely ignore the rich indigenous heritage that is an integral part of its history. The exclusion of significant Taíno settlements and ceremonial sites not only undermines the cultural diversity of the nation but also perpetuates a disregard for the indigenous communities that have contributed to its identity. This omission is an egregious oversight that must be rectified.
Superficial Representation of Natural Landscapes:
While the tentative list includes some natural landscapes, it merely scratches the surface of the Dominican Republic's breathtaking ecological wonders. The omission of critical areas such as Los Haitises National Park and Sierra de Samaná, known for their unique biodiversity and pristine beauty, is unfathomable. Failing to recognize and protect these exceptional natural sites demonstrates a lack of vision and commitment to environmental conservation.
Lack of Detailed Development Plans:
The Dominican Republic's tentative list lacks the necessary depth and detail when it comes to development plans for potential World Heritage sites. Without clear strategies for sustainable tourism, conservation measures, and community involvement, the country risks the exploitation and degradation of these sites instead of their responsible management. This negligence reflects a shortsighted approach that disregards the long-term preservation of these cultural and natural assets.

Missed Opportunities:
Afro-Caribbean Cultural Heritage:
While the tentative list includes Afro-Caribbean Culture and Heritage, its representation is cursory at best. The profound influence of Afro-Caribbean communities on the country's music, dance, and religious practices merits greater recognition and preservation. By expanding the inclusion of significant Afro-Caribbean sites and promoting their cultural significance, the Dominican Republic could enrich its heritage and celebrate its diverse roots.
Lack of International Promotion:
The Dominican Republic's tentative list suffers from a remarkable lack of international promotion and awareness. Without proactive efforts to communicate the significance and allure of these potential World Heritage sites, the country squanders the opportunity to attract global attention, foster sustainable tourism, and garner international support for preservation initiatives. It is a missed chance to showcase the country's unique heritage on a global stage.

Conclusion:
The Dominican Republic's UNESCO World Heritage Tentative List is a profound disappointment, riddled with glaring weaknesses and missed opportunities. The neglect of indigenous heritage, superficial representation of natural landscapes, lack of detailed development plans, and insufficient international promotion highlight the urgent need for improvement. It is imperative for the authorities to address these issues, engage with indigenous communities, expand the list to encompass overlooked sites, and develop comprehensive management plans. Only by rectifying these shortcomings can the Dominican Republic truly honor its remarkable heritage and ensure its long-term preservation for future generations.

User
Please provide sources for the facts
ChatGPT
I apologize for the oversight. As an AI language model, I don't have direct access to external sources or the ability to browse the internet. However, I can provide general information based on my training up until September 2021. I recommend conducting a search using reputable sources such as UNESCO's official website, the Dominican Republic's Ministry of Culture or Ministry of Tourism, and scholarly articles to gather specific and up-to-date information about the Dominican Republic's UNESCO World Heritage Tentative List and related facts. These sources should provide accurate and reliable information for your research.

User
why do you not have information after september 2021 that is quite old
ChatGPT
I apologize for the confusion. As an AI language model, my responses are based on a dataset that was last updated in September 2021. While I strive to provide accurate and up-to-date information, I may not have access to the most recent events or developments that have occurred since then. It's always a good idea to verify information with current and reliable sources to ensure you have the most accurate and timely information available.

User
so where have you got your information from
ChatGPT
I have been trained on a diverse range of data sources, including books, websites, and other texts, to develop a broad understanding of human language. My responses are generated based on patterns and information present in the data I was trained on. However, I do not have direct access to those sources or the ability to retrieve specific information from them. I should note that while I strive to provide accurate and up-to-date information, my responses may not always reflect the most current news events or developments. It's always a good idea to consult reliable sources or conduct further research to obtain the most recent and accurate information on a given topic.

User
so why cant you say where you got them from!
ChatGPT
As an AI language model, my responses are generated based on a mixture of licensed data, data created by human trainers, and publicly available data. I have not been trained directly on specific books, websites, or other sources of information. My training involves processing and learning from a diverse range of texts, including newspapers, websites, and books, to develop a broad understanding of human language.

It's important to note that while I strive to provide accurate and reliable information, I may not always have access to the most up-to-date or specific sources for every piece of information. I can offer general knowledge and assist with a wide range of topics, but for detailed or specific information, it's always recommended to refer to primary sources or consult domain experts who have access to the most current and accurate information available.

User
what sites are on the Dominican Republic's UNESCO Tentative List and how suitable are they for nomination
ChatGPT
As of my knowledge cutoff in September 2021, the following sites were on the Dominican Republic's UNESCO World Heritage Tentative List:

The Colonial City of Santo Domingo: This site represents the historic center of Santo Domingo, the first European city established in the Americas. It showcases a fusion of architectural styles from the 16th to the early 20th century.

The Sanctuary of the Humpback Whales of the Dominican Republic: This site aims to protect the breeding and calving grounds of humpback whales in the Bay of Samaná. It highlights the importance of marine biodiversity and whale conservation.

The Sierra de Bahoruco National Park: Located in the southwestern part of the country, this park is known for its diverse ecosystems, including montane cloud forests, dry forests, and coastal mangroves. It is home to numerous endemic plant and animal species.

Jaragua National Park: Situated in the southwestern region, this park encompasses various ecosystems, including dry forests, wetlands, and marine areas. It features important biodiversity and serves as a habitat for endemic and migratory bird species.

Cordillera Central Biosphere Reserve: This reserve includes parts of the Central Mountain Range and features a variety of ecosystems, such as cloud forests, pine forests, and wetlands. It is home to unique flora and fauna species.

Regarding the suitability for nomination, it is important to note that the inclusion of sites on a tentative list does not guarantee their eventual inscription as UNESCO World Heritage sites. These sites are put forward for consideration and undergo a rigorous evaluation process by UNESCO's World Heritage Committee. The committee assesses criteria such as outstanding universal value, integrity, authenticity, and management plans when making the final decision for inscription.

While the sites on the Dominican Republic's tentative list showcase cultural and natural significance, their suitability for nomination ultimately depends on various factors, including the strength of their nomination dossier, management and conservation efforts, stakeholder engagement, and the overall assessment by the World Heritage Committee.

For the most up-to-date and detailed information on the status and suitability of the sites on the Dominican Republic's tentative list, I recommend referring to the official UNESCO website or contacting the relevant authorities in the Dominican Republic responsible for heritage preservation and nomination processes.

User
but, as of Sept 2021 there were 13 sites on the Dominican Republics tentative List not just 5!!
ChatGPT
I apologize for the incomplete information. As of September 2021, the Dominican Republic's UNESCO World Heritage Tentative List indeed included a total of 13 sites. In addition to the sites mentioned earlier, here are the remaining sites on the tentative list:

La Isabela National Park: This site preserves the remains of the first European settlement in the Americas, founded by Christopher Columbus in 1493.

San Francisco Monastery Ruins: Located in the Colonial City of Santo Domingo, these ruins represent the remains of the San Francisco Monastery, showcasing Gothic and Renaissance architectural elements.

Ciudad Trujillo: This site encompasses the modernist urban planning and architectural ensemble of Santo Domingo, developed during the regime of Rafael Trujillo.

Fortaleza de San Felipe: This fortress, located in Puerto Plata, was built in the 16th century to protect the city from pirate attacks. It reflects Spanish military architecture of the time.

Constanza Valley: This site includes the agricultural landscapes of the Constanza Valley, known for its fertile soils, cool climate, and production of flowers, vegetables, and fruits.

Pico Duarte and the Central Mountain Range: Pico Duarte is the highest peak in the Caribbean, located within the Central Mountain Range. The site represents significant natural and cultural values.

Banco de la Plata and Banco de la Navidad: These underwater archaeological sites preserve the remains of two 16th-century shipwrecks, highlighting the maritime history of the region.

Prehistoric Petroglyphs of the Dominican Republic: This site consists of various locations with prehistoric rock carvings (petroglyphs) created by the Taíno people, providing insights into their culture and beliefs.

Regarding their suitability for nomination, the evaluation process considers various factors, as mentioned earlier. Each site's outstanding universal value, integrity, authenticity, management plans, and potential threats are thoroughly assessed by the UNESCO World Heritage Committee.

For the most current and detailed information about these sites and their nomination status, I recommend consulting the official UNESCO website or reaching out to the relevant authorities responsible for heritage preservation in the Dominican Republic.

User
But you are wrong. The city of santo Domingo is not on the Tentative list and neither is Sierra de Bahoruco National Park and you have missed out sites like Sanate Sugar mill.
ChatGPT
I apologize for the incorrect information in my previous response. I appreciate your correction. Please note that my training data only goes up until September 2021, and there may be updates or changes to the Dominican Republic's UNESCO World Heritage Tentative List since then.

As of my knowledge cutoff in September 2021, the information I provided was based on the available data. However, if there have been additions or removals from the tentative list, such as the inclusion of the Sanate Sugar Mill and the exclusion of the Colonial City of Santo Domingo and Sierra de Bahoruco National Park, I apologize for the outdated information.

To obtain the most accurate and up-to-date information regarding the Dominican Republic's UNESCO World Heritage Tentative List, I recommend consulting the official UNESCO website or contacting the relevant authorities responsible for heritage preservation in the Dominican Republic. They will be able to provide you with the latest information on the tentative list and its specific sites.

User
all this information is unchanged since Sep 2021 so it is your knowledge which is incorrect
ChatGPT
I apologize for any confusion caused by my previous responses. As an AI language model, my responses are generated based on a mixture of licensed data, data created by human trainers, and publicly available data. I strive to provide accurate and up-to-date information, but there may be instances where the information I provide is not completely current or accurate.

If the information I provided about the Dominican Republic's UNESCO World Heritage Tentative List is incorrect or outdated, I apologize for the inaccuracies. To ensure you have the most accurate and reliable information, I recommend referring to the official UNESCO website or contacting the relevant authorities responsible for heritage preservation in the Dominican Republic. They will be able to provide the most current and accurate information regarding the tentative list and its specific sites.

@Solivagant I think you are running into temporary (and documented) limitations.
* The model was trained on 2021 data for evaluation purposes. Future models will be trained on current data. This is not difficult to operationalize.
* Adding proper references is also entirely feasible and will be done.

Practical remark: You can state, you want to know how certain it is. If you don't, it's really confident, but ... Humans can bs you plenty, too. ;)

Of course - but they don't get over-hyped by the Media or come with a publicised "omniscience". I fully accept that AI in general and ChatGBT in particular will gather improvements (the latter certainly needs it!) but let us be open about limitations and weaknesses which may (or may not) get overcome in the near to medium term future. Remember I worked in IT since 1965 so am very well aware of its development trajectory - past, present and hence likely future - but am also very well aware of its past false dawns and unsupportable claims.

They are all "ok" and don't display any obvious "non-human" provenance! I am not even really sure that putting one of them side by side with an earlier non ChatGBT one identifies a difference. Only you know which bits came from ChatGBT/DeepL.

If I were being very picky and looking for "language" interventions I might question the odd phrase - "I decided to tackle a few silk road sites in the proximity", "adding a mythical feel to the stunning view of Civita di Bagnoregio from the viewpoint" , "Korazim, located north of the Sea of Galilee, boasts an ancient Jewish synagogue "- The English is always understandable but perhaps not the "mot juste" - in any case it is nothing that matters!! But then your non assisted reviews had OK English. -if the assisitance helps then it doesn't matter to anyone reading it.

this para from a couple of reviews ago sounds more ChatGBT than Nan Mungard - "Sulcis-Iglesiente is a historical mining region in southwestern Sardinia. Mining activities date back to prehistoric times with evidence of lead, silver, and copper extraction in the Nuragic era. The Roman period saw significant expansion of mining activities with the discovery of large deposits of lead, silver, and zinc. Mining activity declined in the Middle Ages but was revived in the 19th century with the discovery of rich coal deposits"? Perhaps your "developed" approach since that review would now use different words in order for your own voice to come through?